F L O W I N J E C T I O N T E C H N I Q U E S F O R T H E D E T E R M I N A T I O N O F T R A C E E L E M E N T S IN S E A W A T E R
Yury A. Zolotov, lilliya
K.
Shpigun, a n d Irina Ya. Kolotyrkina Kurnakov Institute of General a n d Inorganic Chemistry, Russian A c a d e m yof Sciences, leninsky Prospect 31, M o s c o w 117907, R U S S I A
T h e content and distribution of trace components in the marine environment is of in‑
terest from m a n y points of view. In particular, concentration profiles of s o m e elements in sea water and in interstitial waters of sea‑bottom sediments are found to be caused by the important processes operating within the ocean.
After a sea water sample has been taken, it remains to be biologically active system because of bacteria and plankton continue working on take up and excrete material. Thus, m u c h analytical work must be performed immediately after sampling. In this connection, analytical methods that are rapid, less prone to contamination and m o r e sensitive are re‑
quired. Obviously, automatic chemical analysis can be vital for m a n y marine hydrochemical studies, because analytical results m a y be available within a short period of time.
Flow Injection Analysi s (FIA) is k n o w n to be one of the most promising analytical methodologies for automation of routine wet chemical procedures and for near‑real time process monitoring
111.
lt allows high sample throughputs and provides good reproduc‑ibility of the m e asurements. It w a s recognised that FIA offers m a n y advantages for m a ‑ rine chemical research
12 ‑61.
O u r interest was focused o n the design of flow injection (Fl) techniques, available for shipboard determining trace amounts of transition metals and micronutrients owing to speed of analysis, on‑line sample pre‑treatment and reliable analytical characteristics. In order to provide the full potential of F I A m ethodology for detection low, sometimes ex‑
tremely low, concentration of elements in sea water, however, very sensitive and selec‑
tive analytical reactions or high‑performance flow‑through detectors are required.
It
can be achieved in different w ays. but the use of simple and inexpensive spectrophotometric and electrochemical detectors is the most convenient. For thi s reason , w e developed the Fl m ethods based o n the following principles17 ‑
12I:
I) catalytic effects of m etal ions in h o mogeneous oxidati o n reactio ns of so m e organic combination of in‑valve so rption or gas‑diffusion preconcentration and separation of an analyte followed by potentiometric or spccrophotometric detection of its concentration in the eluate zone ; 3) use of in‑valve electrochemical sensor unit containing a chemically modified carbon com‑
posite electrode (C CE).
A s u m m a r y of the FI techniques developed for determining M n (II ), Fe (Ill), C o (11), Transactions of The Research Institute of (21)
Oceanochemistry Vol. 10, No. I, April. 1 9 9 7
Z n (II) and A g (I) at natural levels in sea water is presented in Table I. It should be m en‑
tioned that all these metals, especially M n (II) and Fe (III), were known to be very impor‑
tant from the geochemical point of view because hydrothermal activity can be recognised by indicating their anomalous concentration in deep sea water 113 ‑ 151.
Table 1. Features of Fl system s proposed for determining som e transition metals in sea water
Species D etection type In‑valve Range of L O O R S D Sampling lnnovertor cone./ n M / n M ／％ rate/ h・1
M n(II) Spectrophot.det based on C olumn with 0.2 . ‑300 0.04 5 ‑8 4 ‑12 catalytic reaction N ,N‑ DETATA ‑
diethyl-aniline-KIO~ sorbent
Fe(III) Reversed FI spcctrophot. 2‑ 1000 0.5 2 ‑ 8 60 det. based on catalytic
reaction N ,N‑dial kyl‑p‑
phenylcndiamine ‑H ,O,
Co(II) Spec1ropho1.de1. based on 0.2 ‑50.0 0.02 3 ‑9 50 catalytic reaction N ,N‑
diethyl‑p‑phenilendiaminc
‑Tiron ‑H ,C）、
Zn(II) Spectrophot.det based on C olumn with . 5 ‑300 0.8 3 ‑ 6 3 ‑ 10 complex form ation w ith Dowe:x トX 8
X ylenol Orange res111
Ag([) A nodic stripping Electrochcm ‑ 0.1 ‑ 10.0 0.05 ‑l‑ ‑10 2 ‑ 6 voltam metry on thiacrown 1cal sensor
modified C CE unil
The schematic diagram of the Fl catalytic system proposed for shipboard determining ultralow concentration of dissolved cobalt in sea w ater is shown in Fig. 1. It was based on the catalytic effect of cobalt (II) in the co‑oxidation of N ,N'‑diethyl‑p‑phenylendiarnine and tiron by hydrogen peroxide
I
101 . T he catalytic activity of cobalt (II) was found to be significantly enhanced by the presence of sea water conservative m acrocom ponents, es‑pecially by calcium ions, improving the sensitivity of cobalt determination in sea water.
T he c o mparatively weak basic m edium (pH 8.7 ‑ 9.0) of the reaction and its relative
freedom from co‑existing ions allowed the direct analysis of sea water to be effectively conducted without any preliminary steps.
200.1°1 S e a water
P u m p
I
6 0 cm/O.? m m 2 2 0 cm/0.7 m m 6 0 cm/0.7 m mヽ
． ，
Carrier
Reagents Oxidant -~:
1.5
0.8 0.6ml/min
．
w
Fig. 1 Schematic diagram of the Fl manifold used for determining cobalt in sea water
Analysis of North Atlantic Smface Sea Water (NASS‑2) and Coastal Atlantic Sea Water (CASS‑2) standards gave cobalt concentrations within the accepted ranges, conforming the accuracy of the proposed technique (Table 2). For comparison, s o m e samples of arti‑
ficial hydrothermal solutions were also analysed by the n e w FI method and by ICP‑MS.
T h e results are also shown in Table 2 and indicate that there were no significant differ‑
ences a m o n g the values obtained by both methods.
Table 2. Results of the determining cobalt (II) in sea water samples ( P
=
0.95)Sample n Fl method
NASS‑2* 4 3.1 土0.5
CASS‑2* 14 27.1
士
0.4Artificial hydrothermal 5 160
士
7solutions
，
200土55 175 土5
(*National Research Council , Canada) Transactions of The Research Institute of (23) Oceanochemistry Vol. 10, No. I, April. 1 9 9 7
Found Co/ ng 1‑1
R S D I C P ‑ M C Certified values
／％ method
10.4 4
士
l2.8 25 土6
3.4 151 土5 2.7 210
士
6 2.5 165士
5Figure. 2 illustrares the Fl system used for the determination of zinc in sea water. As can be seen, it includes a mini column containing a sorbing resin for in‑valve preconcentration of zinc from acidified samples. The spectrophotometric quantification of zinc in the eluate zone was based on the reaction of zinc with triphenylmethane reagent Xylenol Orange.
P u m p l FIitered
Acidified Sea water
紐一⑬
P u m p 2
Jection Valve
一 0.8 2.0 w
1~
(ml/min)
込 Reagent (Xylenol Orange,0.05%.,pHS.7)
)Carrier‑eluent { 2 M NaCl)
Fig. 2 Schematic diagram of the Fl manifold used for determining zinc in sea water
The best selectivity enhancement was achieved by using a strongly basic anion‑ex‑
changer Dowex I‑X8 1111. It was explained by the differences in the stability of the chloro complexes of zinc and of other metals as well as in its preferential distribution coefficient for zinc in relation to potential interfering metals. This procedure also im‑
proved the sensitivity of analysis because the eluent solution (2M NaCl) was compatible with the optimal reaction conditions required for the detection of zinc (pH 5.7). The load time could be varied to concentrate lower or higher expected metal concentration. The accuracy of the method was verified by recovering added amounts of zinc in the artificial sea water samples and by the analysis of CASS‑2 standard reference sea waters (Table 3).
T o demonstrate the applicability of the proposed method to sea water analysis, some Black
Sea water samples were also analysed for zinc. The recovery of zinc which was added to
the sea water samples at concentrations of I O ‑ I 5 0 n M was 96
% ‑I 06
% ,indicating no
significant inte1ference.
Table 3. Recovery test analysis of sea water samples for zinc (n = 6, P = 0.95)
Sample Concentration of Zn/ n M R S D R
A d d e d Found ／％ ／％
Artificial 10 15 土 l 6.3 106
Sea water 3 0 31 土2 5.9 105
(S=34
協）
75 73士
4 4.2 9 6150 153 土5 3.3 102
Black Sea water
゜
12士
l 7.8(S=l6
協）
2 0 33士
2 5.7 1054 0 52
士
2 3.6 100CASS‑2 30 土2 2 9
士
2 6.0 9 7A novel Fl technique with in‑valve electrochemical sensor configuration for determin‑
ing silver by differential‑pulse anodic stripping voltammetry was developed. In order to obtain a method for ultratrace silver determination in sea water which would be easier to use on board ship, a flow‑through electrochemical cell containing a macrocycle modified silica carrier‑carbon composite electrode was fabricated and incorporated into a manifold instead of the sampling loop, as it is shown in Fig. 3. T h e binding of silver by covalently immobilised dibenzo‑18‑tetrathiacrown‑6 w a s established to be very effective for preconcentration of silver from flowing solution
I
161.Silver was selectively collected on the electrode surface at a constant potential ‑of 0.3
V. At the adsorptive collection time of 3 0 min, the electrode shows an oxidation current
response proportional to the silver concentration in the range from 0.1 to 1.0 n M . Table 4 shows the data of the recovery experiments obtained for standard solutions prepared with the acidified Ag‑free sea water, indicating good accuracy and reproducibility of the re‑suits obtained (Table 4). It w a s stated that the developed FI sensor system can detect lower than 0.1 n M of silver.
Transactions of The Research Institute of (25) Oceanochemistry Vol. 10, No. 1, April. 1 9 9 7
1，瓜
6
5
4
3
2
2 M N&CIO,.
ー
/p=(‑0.22
土0.04) + (5.92
土0.06) [Ag)
゜
0.0 0.2 0.4 0.6 0.8 1.0C, n M
l.2Fig. 3 Schematic diagram of the Fl sensor system for ultra‑trace determination of silver
Table
4.Recovery of si lver from sea water (n =
6,P =
0.95)Concentration ofsilver/ n M R S D R
Added found
／％ ／％0.10 0.11
士
0.01 9.4 1100.20 0.19
士
0.02 7.1 950.40 0.41 土0.02 4 .7 103
0.80 0.80土0.03 3.5 100
1.00 1.00 土0.04 3.7 100
The proposed FI techniques were successfully applied to individual determination of dissolved metals in sea water samples collected from at and above the hydrothe
「mal vents in the Manus Basin in Pacific Ocean. Ye11ical profiles of dissolved manganese concentra‑
tion in sea water obtained at the stations are presented in Fig. 4. These data give evidence
of the existence of active hydrothermal springs in this region.
Depth, m
0
￨ ： I
l ,
!
t‑
• I
1 :
i
I
□ 口 ：
0 0 l 5f10 2 G O OM n , ng/1 Fig. 4 Selected manganese profiles in the deep sea water columns of the Manus Basin
M n , ng/1
Depth, m
1:/ ：
／
寧[
i/ l '
.トグ / ノ）イI斤 面 ア が
＇
25OO L .‑ ‑
， ；
C 5 0 0 1 0 0 0
．．
Several sensitive techniques for determining important micronutrients and sulphur forms in deep sea water and interstitial water of bottom sediments were also developed.
Table 5 shows performance data of Fl systems for determining dissolved ortho‑phos‑
phates and silicates by using t w o different types of detection. T h e both methods were based on the well‑known complexation reaction of these components with acidic a m m o ‑ nium molybdate reagent. T h e reaction w a s carried out by using in‑valve reactor unit.
Then, the molybdophosphate or molybdosilicate anions formed in the injector manifold were injected into the carrier stream. In case of amperometric detection, a carrent signal was measured that caused by reduction of heteropolymolybdates at the carbon electrode at zero potential.
T h e data obtained by this simple and sensitive method in analysis of the artificial sea water samples containing ortho‑phosphate and silicate ions are given in Table 6.
Figure. 5 shows the schematic diagram of the Fl system with in‑valve gas‑diffusion configuration and intermitted pumping. It should be mentioned that utilisation of G a s Diffusion Flow Injection Analysis has been demonstrated as an effective technique to increase selectivity and sensitivity for an analytical method
1171.
In the present case, the system allowed the satisfactory determining sulphide and a m m o n i u m ions in sea water at a very low level (< Ir1M).
Table 7 demonstrates the features of these systems with using spectrophotometric de‑
tection and potentiometric detection by membrane ion‑selective electrodes (ISEs).
Transactions of The Research Institute of (27)
Oceanochemistry Vol. 10, No. 1, April. 1 9 9 7
Table 5. Features of the proposed FI systems for determining ortho‑phosphate and silicate in sea water by using two different types of detection
Element Type of FI system Linear cone. range L O O Sampling rate
determined IμM
加
1M /h・lReversed FI system
Ortho‑phosphates based on spectra‑ I ‑ 20 0.2 6 0 ‑ 7 0 photometric detection
FI system based on
amperometric detection 0.1 ‑ IO 0.05 6 0 at a carbon electrode
Reversed Fl system
Silicates based on spectro‑ 5 ‑ 120 3.0 9 0
photometric detection FI system based on
amperometric detection I ‑ 5 0 0.3 6 0 at a carbon electrode
Table 6. Analysis of sea water samples for silicates and phosphates* (n
=
IO, P=
0.95)A m o u n t taken/11M A m o u n t found/1tM
Silicate Phosphate Silicate Phosphate
0.0 0.00 6.4土0.1 0.12 土0.01
5.0 0.10 11.4
士
0.1 0.21 土0.0210.0 0.2 0 16.5 土0.1 0.33
士
0.0225.0 0.30 31.3
士
0.2 0.40土0.034 0.0 0.40 46.5 土0.3 0.51 土0.04
(* Sample volume 300 ml)
Inlet. for
Sea water
Sample・
Pump, 1.2
Donor●tream I 0,8
Acc::eptor P u m p 2 u・tream
后 コ
Reagent 1
‑‑‑‑・ ‑‑‑‑
Reagent 2
ー
ヽ
ヽ
ーJ
一
ヽ一
︑ 一
し .
d
▼
, ・
4,
一
ー︐
ヽ ヽ
︐
ヽ
一 一
ヽ 4゜
． ‑
0.4‑
---•—• ‑‑‑‑‑
(ml/min)
Fig. 5 Schematic diagram of the FI system with in‑valve gas‑diffusion configuration used for determination of trace sulphide and ammonium ions.
Table 7. Features of the proposed FI systems for determining sulphide and ammonium ions in sea water by using two different types of detection (n
=
6 , P=
0 .95)Sulphide Ion A m m o n i u m Ion
S pectrophoto‑ Potentiometric S pectrophoto‑ Potentiometric Parameter metric detection detection with metric detection detection with
S2‑ ‑ ISE N H / ‑ ISE
Donor 0.2 M H̲2 S O 4 0.5 M H ̲ 3 P O 4 0.1 M N a O H 0.1 M N a O H solution
Acceptor 2 .5 m M ‑ N a O H 2 . 5 m M ‑ N a O H Phenol R e d 0 .0 2 M TRIS‑
solution (pH
6.8)
H C I (pH 7 .5)Measuring I ‑ 20 5 ‑ 5 0 0.5 ‑ IO 2 ‑ 100
range/ μM (10‑ 200)* (50 ‑ 500)* (5 ‑ I 00)* (20 ‑ 1000)*
R S D / % 0.4 ‑ 1.0 0.7 ‑4.0 0.5 ‑ 2.0 0.2 ‑ 3.0
L O D I
瓜
0 .25 1.0 0.07 1.0T L.on L. = lo min, （＊） without preconcentrat1on;
Transactions of The Research Institute of (29) Oceanochemistry Vol. 10, No. 1, April. 1 9 9 7
T h e systems were used for routine determining sulphide and ammonium‑nitrogen in sea water samples during two cruises of the research vessel "Academik Boris Petrov".
T h e distribution of dissolved oxygen, p H and redox potential values, total alkalinity, phos‑
phates, nitrates, iron and manganese in Black Sea and Kara Sea were also investigated. In particular, the vertical distribution study of dissolved hydrogen sulphide and a m m o n i u m ‑ nitrogen concentrations in and near the oxic‑anoxic interface zone (to depth as 400 m ) in the Black Sea was carried out (l 9 9 4 Black Sea Expedition of "Akademik Boris PetroY").
T h e presence of sulphide within 150 m of the surface was found. A s can be seen from Fig. 6a, the overlap of the dissolved oxygen and sulphide was observed in the central part of the basin between I 0 0 and 110 m w h e n dissolved sulphide and oxygen coexisted at trace concentration.
T h e vertical profiles of dissolved manganese (II) and iron (Ill) as well as redox poten‑
tial values obtained at this station are shown in Fig. 6b.
(a) D e p t h , Ill
゜
O x y g e n , p p m
2 4 6
I
8
] 20 5
0 5 05 0 50 5 05 0 50 5 0 50 5 00 0 40 0 23 34 45 56 67 78 89 91 01 01 11 11 21 41 7E 50 2 3
ヽ
O x y g e n
‑‑‑‑‑‑‑‑‑ J
ー 一
lntorfaco
S u l p h i d e
6 0 8 0 1 0 0 S u l p h l d e , ̲ p M
1 2 0 1 4 0
(b) Depth~ m
゜
Oeplb,m
゜
︐
50 10 01 50
20 0 25 03 00 35 0 20 0
＞
ー
＇
ー10 0
／
a
・1
｀
ー
＇
/
e
︒
: 2 0 0 [
．
100／ ノ I ::
1/\~L~
0 5 10 1' 20 25
c ,,. ppb C、,.・ppb
Fig.6 Vertical profiles across o、¥ic‑anoxic interface at one of the station in the central part of the
b
゜
p
4
゜ p
r
゜ f e
．
f
゜ r
p a
c
． t r
e
＞
d te
c
e
e
••
,．
・ ・
・
1 .. , . .. ..
.
s
als
] 訓 訓 馴
a
ー︐
．．
︐．
︐
．．
．
︐
．
．
．
n e
9 . . . . 9 . .
g s t '
•••
﹄
I. A
．岳
r
ド1 E ,
．．．
．
．．
． p . :
̀ . 0•••••
︐
．
．．
．
s a
心
i . I i
ー'1••
r
．．
． , .,．F
., .5 ,＂
．
．
．．
9
＂.
．．
．
？ .
︑
．．
F
•.
9 •
•••
︑．
〇
( i
•••
• 9
.. .. . ,
..••.••••
l. . ..
••••.•••.••••••••
4
.
．︐
9 .
a l k 4
＇
:； ・
・ ・
・ ・
n o •
••••
>
,
．．
．．．
︐． ,
．
．．
．
c i j
ー ・
1 9 ,'． ー
9'
t •I
yP
﹄ 叫
ビ J
・ ︐ 札 ． ー ド
・ ・
・ ・ ペ ．
＇"
・
‑
・
・‑
0.i.‘:
·r:•
9
s a
加
o f t
̀ 0
咲 ・
．
••
8 00
州 ．
． ．
．
e 7 n e o r
ua t 5
g d o
．
e o
n
F
．
a
t
Depth, m
6
゜
15 2 0 3 0 .60 100 2 0 0
0.05
Phoaphate
、
μM0.16 0.2 0.26
0.1 0.3
3 0 0 7
7.6 e
p H
8.6 g
Fig. 7 Typical analytical signals and selected profile of dissolved phosphates at one of the station in K A R A Sea
In conclusion, the Fl techniques described here have advantages for shipboard use over conventional methods in that. T h e y are simple and inexpensive, have increased sen‑
sitivity and selectivity caused of kinetic discrimination effects, and can provide geochemi‑
cally consistent analytical results with good reproducibility and high sampling rate. These techniques are widely applicable to a variety of hydrochemical studies.
References.
I..I.Ruzicka and E.H.Hansen: "Flow Injection Analysis" , N e w York, John Wiley & Sons, 1981, p.207.
2. Y u.A Zolotov, L.K. Shpigun, I.Ya.Kolotyrkina, O .V.Bazanova, and E.A. Novikov: Anal. Chim.
Acta, 1987, 200, 21.
3. L.K.Shpigun : Zh. Analit.Khin. , 1990, 45, I 045.
4. T.P Chapin, K .M.Johnson, and K . H.Coale: Anal,Chim.Acta, 1991, 249, 469.
5. J.A.Resing and M.J .Mottl : Anal.Chem., 1992, 64, 2682.
6. K.N.Andrew, N..I .B lundeil , D .Pricc and P.J.Worsfold: Anal.Chem., 1994, 66, 916A.
7. L.K.Shpigun, I.Ya.K olotyrkina, and Yu.A.Zolotov: Anal. Chim. Acta, 1992, 261 , 307.
8. I.Ya.Kolotyrkina, L.K.Shpigun, Y u.A .Zolotov, and G.I. T sysin : A nalyst, 1991 , 116, 707.
9. I.Ya.Kolotyrkina, L.K.Shpigun, Yu.A.Zolotov , and A. M alahoff : Analyst, 1995 , 120,20 1.
10.A.Malahoff, I.Ya.Kolotyrkina and L.K.Shpigun: Analyst, 1996, 121 , 1037.
11. L.K.Shpigun, G o W ei, and P.M . Kamil ova: Abstr. of Russian Symposium on Flow Analysis, M o s c o w, Russia, 1994, p.52.
Transactions of T h e Research Institute of Oceanochemistry Vol. 10, No. I, April. 1 9 9 7
(3 1)
12.L.K.Shpigun : Abstr. of Pacific Congress on Marine Science and Technology, Honolulu, U S A . 1996, p.77.
13 .G .A.Knauer, J.Y.Martin, and R .M .Gordon: Nature, 1982, 297, 49.
14.Y.Bouquet, U von Stackelberg and J.L.Chalou: Nature, 1991 , 3 4 9,778.
15 . P.N. Sedvick, G . M . McMurtry, and J.D. Macdougall : Geochim. Cosmochim. Acta, 1992, 56, 3463.
16.E.I.Troyansky, M.S.Pogosyan, N .M.Samoshina, G.I.Nikishin, V.V.Samoshin, L.K.Shpigun, N .E.
Kopytova, and P.M.Kamilova: Mendeleev C o m m u n., 1996, 1, 9.
17. R .Nakata, T.Kawamura, H.Sakashita, and A .Nitta: Anal.Chim.Acta, 1988, 208, 81.
［和文要旨］
フロー分析法による海水の微量元素の定量
クルナコフ一般化学無機化学研究所
Yu. A
ゾロトフ 、L. K.
シピグン 、I.ya. コロチルキーナ海洋の本性を化学的に理解するには、分析化学を発展させることと分析機器を 進歩させることが不可欠である。観測船上での測定に各種のフロー分析法を採用 することを目的として、フローインジェクション分析
(FIA)
法を詳しく研究し た。FIA
法は丈夫であることの他にも様々な利点を持つため、海が荒れていても 採水後すぐに誤差の少ない分析結果を得ることができる。そして、海水の微量元 素の定量や観測においてFIA
法の特性を十分に生かすには感度及ひ選択性の高い 化学反応とフロースルー型検出器が必要である。本研究では、海水の微量溶存成分である
Mn
、Fe
、Co
、Zn
を定量するためのFIA
装置を開発した。微量の
Fe
とCo
の定量には、これらの金属イオンが有機物の酸 化反応の触媒となることを利用して、分光光度法により検出を行うFIA
を、また、Mn
とZn
の定量用には、予備濃縮を行うために 、インジェクションバルブにイオ ン交換樹脂を満たした微小カラムを装備したFIA
を開発し た。これらの方法によって、
Loibe
海底火山のPele
噴出孔で採取した海水試料中の溶存態Fe
とMn
の定量に成功した。山頂の熱水噴出孔及ぴその上方で採取した海水は、
Mn
とFe
を 異常に高濃度で含んでいることかわか った。その他、海洋深層水と海洋堆積物の間隙水に含まれている栄養塩と有害成分の 重要なものを定量するための
FIA
を提案する。観測船'Academician Boris
Petrov"
の航海において 、FIA
を使用し 、海水 中の硫化物イオ ンとアンモニア態窒素を自動測定した。黒海と
Kara
海では、 溶存酸素、pH
、酸化還元電位、アル カリ度、リン酸、硝酸の分布の調査も行った 。