Henryk Zbigniew WREMBEL THE DETERMINATION OF MERCURY

Henryk Zbigniew WREMBEL
U niversity of Education in Słupsk
THE DETERMINATION OF MERCURY
AT ULTRAMICROTRACE LEVELS IN WATER
USING THE LOW-PRESSURE RING DISCHARGE
C o n t e n t s : 1. Introduction, 2. Method, 3. M easuring point, 4. A pplication lim its
of the method, 5. E ffect of w ater com ponents on the results, 6. D eterm ination of
m ercury in the B altic Sea w aters, 7. Conclusions; Streszczenie; References.
1. INTRODUCTION
The ability of m ercury to bind w ith plant and anim al proteins favours
accum ulation of th e elem ent in living organisms. The bioconcentration
of m ercury occurs p articu larly efficiently in the m arine food chains.
In fish tissue the concentration of m ercury by accum ulation can be by
3—5 orders of m agnitude higher than th a t in sea w ater. The m ercury
content in m arine p red ato ry fishes can be estim ated from the following
relationship [8] provided th a t th e m ercury level in sea w ater is known:
CF = 0.013 Cw -
0.179,
w here CF is the concentration of m ercury in fish, in mg kg"1, and Cw
is the concentration of m ercury in w ater, in mg d m '3.
The m ercury content in m arine p red ato ry fishes can attain high
values (Fig. 1), even in w aters w ith relatively low concentration of m er­
cury. The bioconcentration of m ercury in fish is m arkedly m ore effi­
cient at higher levels of the elem ent in w ater. M any papers have been
confined to th e problem of assaying low levels of m ercury in w ater.
Those which constituted th e tu rn in g point w ere those by Stock and
Zim erm ann [16], who developed the m icrom etric m ethod for assaying
m ercury, and by H atch and O tt [11] who applied flam eless atomic ab­
sorption spectrom etry for th e determ ination of m ercury in w ater. This
method, together w ith various m odifications of th e procedure, has found
w idespread application to assay m ercury in w ater at u ltram icrotrace
levels (10"8 < C ;. < lO-5 g dm"3) [2, 5, 19]. However, recently intensive
studies have been carried out to assay m ercury in w ater at subm icroOceanology No 13(1980)
Fig. 1. Mercury contemt in m arinę fishes vs.
cancemtration of mercuiry
in sea w ater (after [8]).
10
¿0
60
-3
Hçj co n ce ntra tio n in w a te r C ^ .n g .d rn
20
30
50
70
C»
Rys. 1. Zawartość rtęci
w rybach m orskich w za­
leżności od stężenia rtęci
w morzu (według [8])
trace levels (10*11
C ^-^IO '9 g dm"3) by other m ethods [1, 12, 13], At
these levels the m ethod of flam eless atomic absorption spectrom etry
has found only lim ited application [2 ,14].
This paper shows the possibility of using th e atomic emission spec­
trom etry of a low -pressure ring discharge for assaying m ercury in w a­
ter. The ring discharge utilized in this w ork occurred over th e pressure
range 13 — 1 333 P a (0.1 — 10 Tr), hence u nder the term ,,ring dischar­
ge”, this particu lar kind of discharge is m eant. The restriction of this
term appears to be perm issible, since the high-pressure kind of ring
discharge has commonly been term ed „inductively-coupled plasm a” [7,
9, 18], Owing to im provem ent of the m ethod and exchange of some ac­
cessories, the sensitivity of the m ethod increased and th e detection lim it
of m ercury in w ater changed as com pared w ith earlier resu lts [20, 21],
2. TIIE METHOD
M ercury was assayed in v/ater by m easuring th e intensity of th e Hgl
lines of m ercury selected from the emission spectrum of th e low -pressu­
re plasmoid of th e ring discharge.
Fig. 2. Block diagram of the mercury concentration m easuring point. HF —
h igh-frequency generator; I — inductance coli (exciter); S — spectral lamp: M —
monochromator; Ph — photomiultiiplier; R — recorder
Rys. 2. Schem at blokow y stanow iska pom iarowego służącego do oznaczania
rtęci. HF — generator drgań elektrom agnetycznych, I — cew ka indukcyjna (wzbud­
nik), S — lamtpa spektralna, M — m onochromator, Ph — fotopow ielacz, R — re­
jestrator
To ru n the determ ination, a 1-cm3 sample of the w ater to be analy­
sed was vacuum -dried at 243 K (-30°C) in the discharge cell of th e spec­
tra l lamp. A fter drying the sample, the lamp was evacuated to a pressu­
re of 13 P a (0.1 Tr) followed by switching on the generator of high-frequency electrom agnetic field to cause ignition of the plasm oid of the
ring discharge in th e discharge cell of the spectral lamp. The radiation
of the plasm oid was directed (Fig. 2) into the inlet slit of the m onochro­
m ator. The m ercury spectral line isolated from the radiation of the pla­
smoid was directed to a photom ultiplier and the generated photocurrent
to a recorder cr electrom eter through an am plifier. The m ercury con­
te n t of the sample was read out from the recorder or electrom eter.
The low -pressure ring discharge, som etim es referred to as the
H -type discharge, is generated w hen the discharge cell containing a ra ­
refied gas is placed in the alternating m agnetic field of an inductance
coil (providing an exciter) coupled w ith a high-frequency generator.
The spectrum of the plasm oid of th e low -pressure ring discharge car­
ried out in a spectral lamp filled w ith air at a low ered pressure of the
order of 13 — 1 333 P a (0.1 — 10 Tr) has a rich stru ctu re over the visible
and near-ultrav io let regions [20, 21], The stru ctu re is made up of the
lines of atoms of the elem ents being the com ponents of air and its im ­
purities. In the ring discharge m ercury is excited v ery strongly and in
the spectrum of th e plasm oid all ultim ate lines of the elem ent appear.
W hen the air in the lam p is replaced by helium , the m ercury lines are
strongly amplified.
The content of m ercury was determ ined photom etrically based on
the intensity of photocurrent generated in the photom ultiplier. The m e­
asurem ents w ere ru n for the atomic lines of m ercury, H gl of 253.62 and
435.83 nm. As the la tte r was b etter fitted to the region of the m axim um
sensitivity of the photom ultiplier, it proved to be th e most suitable for
the determ inations. This was particu larly adventageous w hen using a
spectral lamp filled w ith helium as a buffer gas. A part from th e m ea­
surem ents of the intensities of the 253.62 and 435.83-nm lines accompli­
shed by m eans of the photom ultiplier, visual control observations (by
means of a spectroscope) w ere made of the 546.07-nm Hgl line, which
was very strongly excited in the spectral lamp employed.
3. MEASURING POINT
The m easuring point (Fig. 3) was assembled by using generally accessi­
ble analytical and m eansuring instrum entation. A few specially — con­
structed devices provided the exceptions. The appliances consisted
of the following:
(i) S pectral lamp. A specially-constructed electrodeless spectral lamp
was used, the schem atic of which is shown in Fig. 4. It consists of a dis­
charge cell and a connector. The discharge cell was in th e form of a
quartz (Vitreosil) cylindrical cuvette (150 mm lang X 11.5 mm i.d.)
Fig. 3. Schem atic diagram o f th e point used for m easuring th e concentration
of m ercury in ¡sea w ater. HF — generator of electrom agnetic vibrations; I — exci­
ter; O — oil ejector; P — vacuum pump; S — spectral lam p; Tr •— vacuum meter;
Sp — spectroscope; L — quartz lens; M — monochromator; Ph — photom ultiplier;
R — tape recorder; E — electrom eter; VC — digital voltm eter
Rys. 3. Schem at stanow iska pom iarowego użytego do oznaczania rtęci w w o­
dzie m orskiej. HF •— generator drgań elektrom agnetycznych, I — wzbudnik, O —
odrzutnik oleju, P — pompa próżniowa, S — lam pa spektralna, Tr — próżniom ierz, Sp — spektroskop, L — soczewka kw arcow a, M — m onochromator, Ph —
fotopow ielacz, R — rejestrator, E — elektromeitr, VC — w oltom ierz cyfrow y
Fig. 4. Schem atic diagram of the spectral lamp;
1 — valve cutting off the vacuum system ; 2 — con­
n ection w ith the vacuum system ; 3 — connector; 4 —
vacuum joint; 5 — plasm oid of the ring discharge;
6 — inductance coil (exciter); 7 — filter; 8 — valve
for filling the pump
Rys. 4. Schem at użytej lam py spektralnej; 1 —
zawór odcinający układ próżniow y, 2 — połączenie
z układem próżniow ym , 3 — łącznik, 4 — szlif próż­
niow y, 5 — plazm oid w yładow ania pierścieniow ego,
6 — cew ka induktora (wzbudnik), 7 — filtr, S —
zawór służący do napełniania pompy
The connector, equipped w ith a vacuum valve to cut off the lamp from
the vacuum system and another valve to aerate the lamp, was made
from P y rex glass. The discharge cell was attached to the connector by
m eans of a vacuum jonit;
(ii) Spectrom etric system. This consisted of a prism atic m onochro­
m ator (SPM-2, C arl Zeiss in Jena), the spectroscope (UM-2, Unimash)
for visual observation of the 546.07-nm line and the photom ultiplier (M
12 FQS 35, W erk fü r Fernsehelektronik);
(iii) The vacuum system. This consisted of a rotation vacuum pum p
(BL-82, UNITRA), oil ejector (of own construction), and a constant-resistance vacuum m eter (PSO-69, ZOPAP);
(iv) The recording system, which included a tape recorder (K-201,
Carl Zeiss, Jena), an electrom eter (219A, UNITRA)' and a digital m u lti­
m eter (VC-10T, UNITRA);
(v) Feeding system of the spectral lamp. A specially-constructed
high-frequency generator (f = 62 MHz, P out = 8 4 W) was used. The
generator was fed w ith an a.-c. 220-V voltage stabilized to w ithin 0.01%.
The m easurem ents w ere perform ed at a slitw idth of 0.025 mm and
slit height of 20 mm.
4. APPLICATION LIMITS OF THE METHOD
Calibration graphs for th e m easuring system w ere constructed separa­
tely for distilled w ater, artificial Baltic Sea w ater and artificial sea
w ater. A rtificial Baltic Sea w ater (Bz) was prepared by em ploying the
procedures reported in [10, 17], w hereas artificial sea w ater (S2) was
prepared according to the Polish S tandard Artificial sea water [23], In
both kinds of sea w ater the suppression of some m ercury lines was ob­
served. The suppression was relatively strong for the 253.62-nm line.
This was m ostly due to the interaction of radiation v/ith some atoms
of the salts dissolved in the w ater, m ainly w ith chlorine (Cl) and p a r­
tly w ith chlorine m olecules (Cl2) The filling of the spectral lamp
w ith helium resulted in a very strong intensification of th e atomic
lines of m ercury. In th e presence of helium th e intensity of the 435.82-nra line increased considerably. In Fig. 5 a calibration graph is show
for samples of distilled w ater polluted w ith m ercury, constructed for
the 253.62-nm line.
The calibration graph of the m easuring system obeyed th e followig relationship for th e m easuring point and the m ethod employed:
JJ = a C5 l e - kCJ>
where J } is the photocurrent, Cj =10~J g dm"3 is the concentration
of m ercury in w ater, and a, b and k are experim ental constants. The
7 — O ceanologia
i-
-l.'trif c s i .Aj .
..r U . ...
s. b . n
P h o tom u ltiplier
Fig. 5. Calibration graph of the
m easuring system foir distilled waiter
Rys. 5 K rzyw a cechow ania układu
pom iarow ego dia w ody destylow anej
response J j , n A
Fig. G. Calibration graphs of the .me­
asuring .system for th e ranges of u ltramicro and subm icrotrace concentrations
of mercury: D —■ in distilled w ater, B z
— in artificial Baltic Sea waiter, S z —
in artificial sea water.
Ryis. 6. K rzyw e cechow ania układu
pom iarow ego w zakresie uiltramikro i subimikrośladowych .stężeń rtęci: D — w w o ­
dzie destylow anej, B z — w zastępczej
w odzie bałtyckiej, Sz — w w odzie mor­
skiej
coefficient, k, describing self-absorption of the radiation in the plasmoid, am ounted to 0.037. For the u ltra — and subm icrotrace levels
of m ercury in w ater (10_8^ C y. =d0"e and 10'n,^Cy. <10~9 g d m '3, res­
pectively), th e attenuation of the signal due to self-absorption is n e­
gligible (Fig. 5) and consequently the expression for the calibration graph
assumes the form:
Jj = a C 1In the distilled w ater th e values of a=2350 and b = 0.23 w ere ob­
tained over this concentration range of m ercury for th e 253.62-nm
line.
Due to attenuation of the 253.62-nm line by some com ponents of
the sea w ater, the calibration graphs for particu lar kinds of the w ater
differred. The sensitivity of the m ethod was also low er for th e two
kinds of sea w ater than for distilled w ater.
In Fig. 6 calibration graphs of the system are shown over the u l­
tram icro trace concentration range of m ercury for the th ree kinds of
w ater.
The detection limit, CL, was determ ined from the equation reported in [6]:
JL =
Jn +
s j / | '
[tL + t N)
w here J L and J N are th e signals corresponding to th e detection lim it
and noises, respectively; s = m ax. (S i , s#), w h ere sL an d
are s ta n d a rd
deviations of the signal and noise, respectively: t L and t N are respec­
tive critical values of th e t variable of the S tudent distribution.
The determ ination limit, CD, of m ercury in w ater was determ ined
as a concentration at which th e erro r of the m ethod am ounted to 25 per
cent. It w as calculated from th e expression reported in [13]:
J d = Jn +
w here J D and J N are signals corresponding to the detection lim it
and noise, respectively.
Table 1 lists num erical values obtained for the detection and de"
term ination lim its for various kinds of w ater.
Table 1. D etection lim it (C7 ) and determ ination lim it (CD ) of m ercury in various
kinds of w ater (D — distilled water; B . — artificial B altic Sea water; S z — arti­
ficial sea water)
Tab. 1. Granica w ykryw alności (C L ,) i oznaczalności (C D ) rtęci w różnych odmia­
nach w ody (D — woda destylow ana, B j — zastępcza woda bałtycka, S* — zastęp­
cza woda morska)
Kind of w ater
Odmiana wody
CL, mg dm-!
C d , ng dm -3
D
0.001
0.3
Bz
0.01
2.0
0.2
10.0
A series of control determ inations confirm ed the suitability of the
developed m ethod and technique for assaying ultram icro trace am ounts
of m ercury in sea w ater and subm icrotrace am ounts of the elem ent
in distilled w ater. The results of these determ inations are listed in
Table 2.
The m ethod and m easuring technique developed in this w ork en­
abled to shift the determ ination lim it of m ercury in w ater at least
by one order of m agnitude as com pared w ith the flam eless atomic ab­
sorption spectrom etry. By using this m ethod and technique it is possibTable 2. R esults of control m easurem ents in various kinds of w ater (D — distilled
w ater; B . — artificial B altic Sea water; Sz — artificial sea water; SR — relative
standard deviation; n = 20)
Tab. 2. W yniki pom iarów kontrolnych w różnych odmianach w ody (D — w oda
destylow ana, Bz — zastępcza w oda bałtycka,
— zastępcza woda morska, SR —
w zględne odchylenie standardowe, n = 20)
Kind of
w ater
Odmiana
w ody
D
Bz
Sz
Hg added
Dodano Hg
ng dm-5
0.00
100.00
0.0
100.0
0.0
100.0
H g f o u nd
Wylkryto Hg
n g dm-3
0.1
97.6
0.4
92.7
0.5
89.4
S
r
0.09
0.12
0.17
le to assay m ercury in w ater over th e range betw een ultram icro and
submicro trace levels.
5. THE EFFECT OF WATER COMPONENTS ON RESULTS
To determ ine the effect of sea w ater com ponents on th e signal in ten ­
sity, a series of m easurem ents w ere ru n for the 253.62-nm line w ith
populations of samples of nom inally equal m ercury concentrations, p re ­
pared by using solutions containing some com ponents of the artificial
sea w ater only.
Solution Ri contained only sodium compounds which make up
approx. 79.4 p er cent by w eight of all com ponents of artificial sea
w ater. Solution R2 contained th e rem aining com ponents of the w ater,
w ith the exception of sodium compounds, w hich constitute about 20.6
per cent of all components. Solution R 3 com prised heavy m etal n itrates
w hich occur in artificial sea w ater at a level of 0.0005 p er cent.
As regards samples prepared using distilled w ater, suppression of
the signals was observed-in these solutions (Fig. 7). In an extrem e case,
the attenuation of the signal in artificial sea w ater of full composition
am ounted to approx. 65 per cent. In the rem aining solutions, the stron­
gest attenuation was noted in solutions containing chlorine and sodium
compounds. O ther sea w ater com ponents did not affect th e signal intesity appreciably.
Fig. 7. S ignal in ten sities fo rssolutions of various contents of
sea w ater com ponents (norma­
lized to signal 'in distilled w a­
ter) at various concentrations
of m ercury in water.
Rys. 7. W ielkość sygnału dla
roztw orów o różnych zaw artoś­
ciach składników w ody morskiej
(unorm owane do sygnału otrzy­
m yw anego dla w ody destylow a­
nej) przy różnych ¡stężeniach
rtęci w w odzie.
6. DETERMINATION OF MERCURY IN BALTIC SEA WATER
The concentration of m ercury in Baltic Sea w ater was m easured by
this apparatus and m ethod in 1976— 1978. W ater samples w ere taken
in the coastal zones of Kołobrzeg and U stka and in th e region of th e
Słupsk Sandbank (Ławica Słupska). A few samples taken in th e G ulf
T able 3. The content of m ercury in som e B altic Sea regions (C =
= concentration o f m ercury in water); c = m ean concentration of
m ercury in water; S R = relative standard deviation; n = number
of m easurem ents)
Tab. 3. Zawartość rtęci w niektórych obszarach Bałtyku (C — stę­
żenie rtęci w w odzie, c — średnie stężenie rtęci w wodzie, S r —
w zględne odchylenie standartowe, n — liczba pomiarów)
R egion
Rejon
C, ng dim-5
c, ng dm-3
Sr
n
C oastal walterls
(Ustka)
Wody przybrzeżne
20— 180
40
0.75
30
C oastal waiters
(Kołobrzeg)
W ody przybrzeżne
30— 190
45
0.82
30
10—>150
34
0.45
12
120—360
105
0.67
7
Słupsk Sandibank
Ław ica Słupska
G ulf o f Gdańsk
(Sopoit)
Zatoka Gdańska
of G dańsk w ere also analysed. They w ere stored in polyethylene con­
tainers and th e analyses w ere ru n 24 hrs. after sampling. The sam ­
ples w ere acidified w ith n itric acid to pH<[2 and 0.1 g of potassium io­
dide was added per dm 3. N itric acid was added to minimize losses of
m ercury due to sorption on th e w alls of the container, w hereas po­
tassium iodide decreased m igration of m ercury from th e solution to
the atm osphere. The results are shown in Table 3.
There w ere considerable fluctuations in the concentration of m er­
cury in the Baltic Sea w aters. Sim ilar resu lts from other regions of
the Baltic Sea w ere reported in [3], E xtrem al concentrations of m er­
cury in these regions differred by tw o orders of m agnitude. A too
short observation period excluded
th e possibility of revealing the
cyclic nature of variations of the m ercury concentration.
7. CONCLUSIONS
The m easurem ents carried out in this paper confirm ed the suitability
of the m ethod for assaying u ltram icrotrace levels of m ercury in sea
w ater. The m ethod is p articu larly useful in investigations of subtle
processes of exchange of m ercury betw een the solution and walls of
vessels used in analytical procedures, as w ell as betw een th e atm os­
phere and w ater surface. These processes provide the main sources of
vitiation of the results in the analysis of ultram icrotrace am ounts of
m ercury in w ater [22],
The results of the m easurem ents of the concentration of m ercury
in the Baltic Sea w ater showed th a t in the regions studied the m ercury
content ranged from 30 to 60 ng d m '3, occassiona'ily attaining levels of
10 and 100 ng dm"3 (in extrem al cases even more than 350 ng dm"3).
The concentration of m ercury in th e Baltic Sea exhibits seasonal va­
riations.
It w as also found th a t the sensitivity of th e m ethod was largely
affected by proper fitting of the analytical line of th e elem ent to
the m axim um sensitivity range of th e photom ultiplier, as w ell as the
use of proper bufferring gas for filling the lamp. The use of helium
as the bufferring gas increases th e sensitivity of the m ethod in the
case of m ercury.
Wyższa Szkoła Pedagogiczna w Słupsku
OZNACZANIE MIKROŚLADÓW RTĘCI W WODZIE Z UŻYCIEM
NISKOCIŚNIENIOWEGO W YŁADOWANIA PIERŚCIENIOWEGO
Streszczenie
Zdolność rtęci do w iązania się z białkiem sprzyja akum ulow aniu się jej w or­
ganizm ach żyw ych i biokoncentracji rtęci w łańcuchach pokarmowych. Biokoncentracja rtęci przebiega szczególnie efek tyw n ie w wodnych łańcuchach pokarmo­
w ych. W tkankach ryb m oże nastąpić zatężenie rtęci, spowodowane akum ulacją,
naw et o 3—5 rzędów w ielkości w porównaniu z zawartością rtęci w wodzie. P o ­
woduje to konieczność oznaczania naw et bardzo niskich zaw artości rtęci w w o ­
dzie.
W pracy przedstawiono m ożliw ość użycia em isyjnej spektrom etrii n iskociśn ie­
niow ego w yładow ania pierścieniow ego do oznaczania ultram ikrośladów rtęci w
wodzie. Osiągnięta granica oznaczalności w ynosząca dla w ody destylow anej ok.
0,1 ng. dm-3 Hg/H20 , k w alifiku je się m etodę zw łaszcza do badania subtelnych pro­
cesów w ym iany rtęci m iędzy roztworem a otoczeniem , np. m iędzy roztworem
a ściankam i naczyń,
stosow anych w technice analitycznej lub do badania pro­
cesów w ym iany rtęci m iędzy atm osferą i powierzchnią wody. Zjawiska te są
w zakresie ultram ikrośladów głów nym i źródłami błędów oznaczeń rtęci w roz­
tworach wodnych.
Badania w ykonane w wodach bałtyckich w ykazały, że zawartość rtęci, w ob­
jętych badaniami obszarach morza, mieścji s ię naogół w granicach od 30 do
60 ng. dm-J, osiągając czasam i w artości rzędu 10 ng. dm-*, a niekiedy powyżej
100 ng. dm-5, w skrajnych przypadkach naw et powyżej 350 ng. dm-3.
Stwierdzono również, że dla czułości m etody duże znaczenie ma dopasowanie
linii analitycznej oznaczanego pierw iastka do zakresu m aksym alnej czułości w id ­
m owej fotopow ielacza użytego do detekcji prom ieniowania, jak rów nież u życie od­
powiedniego gazu buforowego służącego do napełnienia lam py spektralnej. U życie
helu jako buforu powoduje przy rtęci zw iększenie czułości metody.
REFERENCES
1. B o u d i n G., A n alytical Methods A pp lied to W ärter Pollution , J. Radioanal.
Cheim., 1977, 37, p. 123 — 139.
2. B r e i id e n b a c h A. W., C r o w e R. E., Methods for Chemical Analysis of
W ater and Wastes, U. S. E nvironm ental Protection A gency, W ashington 1974.
3. B r i i g ' i m a n n L., Some R em arks on the Content of M ercury in the Baltic,
Proceedings of the X I Conference of the B altic Oceanographer«, Rostock 1978,
1, p. 249 — 263.
4. B r z e z i ń s k a A., Trace Metals in the W ater of the Gdańsk Basin, Procee-
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
dings of the X I Conference of the B altic Oceanographers, Roistock 1978, p.
264 — 269.
F ö n s t n e r U., M ü . l l e r G., S ch w erm etalle in Flüssen und Seen, Springer,
B erlin 1974.
G a b r i e l R., A General M ethod for Calculating the Detection L im it in Che­
mical Analysis, Anal. Chem., 1970, 42, p. 1439 — 1440.
G a j e w s k i W., (Ed.), Zagadnienia p o d sta w o w e spektraln ej analizy atomo­
w ej, WNT, W arszawa, 1972.
G a r d n e r D., M ercury in Fish and W aters of the Irish Sea and other U nited
K ingdom F ishing Grounds, Nature, 1978, 272, No. 5648, p. 49-51.
G r e e n f i e l d S., M c G e a c h i n H. Ms;D., S m i t h P. B., Plasma Emission
Sources in A nalytical Spectroscopy, Talanta, 1976, 33, p. 1 — 14. .
H o r n e R. A., Marine Ch em istry, W iley, N ew York 1969.
H a t c h W. R., O t t W. L., D eterm ination of S ub-m icrogram Quantities of
Mercury b y Atom ic Absorption S p ectro m e try, Anal. Chern., 1968, 40, p. 2085
— 2087.
L i e s e r K. H., N e i t z e r t V., Determination of Trace Elements in Water
b y N on-destructive Neutron Activation, J. Radioanal. Chem., 1976, 31, p. 397
— 405.
M o e n k e H., A tom spektroskopische Spurenanalyse, A kadem ische Veiri'äSsgesellsch aft, Leipzig 1974.
M u z y k o v G. G., C h u v stv ite ln o st’ opredeleniya rtu ti atomno-absorptsionn y m m e to dom izholodnogo para, Zhurn. Priklad. Spektoroskopii, 1975, 23 (5),
p. 763 — 767.
R e i c h e r t J. K., Spurenan alytik in G ew ässern m it Hilfe der Atomem ission,
A tom absorption und Atmofluoreszenz, Vom W asser, 1975, 40, p. 135 — 149.
S t o c k A., C u c u e l F., Die Verbreitung des Quecksilbers, N aturw iss., 1934,
22, p. 390 — 393.
T r z . o s i ń s k a A., P o d s ta w o w y sk ład jo n o w y w o d y bałtyckiej, Studia i M ate­
riały Oceanologiczne, 1977, 17, p. 165 — 180.
W i n e f o r d n e r J. D., Trace Analysis — S pectroscopic Methods for Elemen­
ts, J. W iley, N ew York 1976.
W i n k l e r H. A., Vorkom m en von Quecksilber in G ew ässern und seine ana­
lytische Erfassung, Chemie Ingenieur Technik, 1975, 47 (16), p. 659 — 666.
W r e m b e l H. Z., Zastosowanie e m isy jn e j analizy w i d m o w e j do w y k r y w a n i a
śladow ych ilości rtęci w wodzie, S tudia i M ateriały Oceanologiczne, 1977, 17
p. 415 — 438.
W r e m b e l H. Z., Z astosowanie o p tyczn ej analizy em isy jn e j do w y k r y w a n ia
śladow ych ilości rtęci w wodzie, Studia i M ateriały Oceanologiczne, 1977, 19,
p. 128 — 135.
W r e m b e l H. Z., The Use of Ring Discharge S p e c tro m e try for Some In ve­
stigations of the Behaviour of Mercury in W ater, P roceedings of the XT Con­
ference of the Oceanographers, Rostock, 1978, 2, p. 679 — 691.
Z astępcza woda morska, Polska Norma PB-66/C -06502, 1966.