AMMONIUM IN ZEOLITIZED TUFFS OF THE KARLOVASSI

Comments

Transcription

AMMONIUM IN ZEOLITIZED TUFFS OF THE KARLOVASSI
Canodian Mineralogbt
Vol. 30, pp.423-430 (1992)
AMMONIUM IN ZEOLITIZED
TUFFSOF THE KARLOVASSIBASIN,
SAMOS, GREECE
ANTHONY HALL
Department of Geologlt, Royal Hollowoy and Bedford New College (University of London),
Egham, Surrey TW20 OEX, U.K.
MICHAEL G. STAMATAKIS
Department of Geology, Notional University of Athens, Ponepistimiopolb, Ano llissio, 157 84, Athens, Greece
ABSTRACT
Miocene tuffs of the Karlovassibasin on the island of Samos,Greece,are largely altered to secondarydiagenetic
silicateassemblages.
Representativesamplesof theserocks have beenanalyzedfor the ammonium ion. Ammonium
was found to be presentin all the zeolitic rocks Q9-579 ppm, mean = 276 ppm), but there are higher levels of
ammonium in those tuffs that are enrichedin authigenicK-feldspar (316-913ppm, mean = 504 ppm). Comparisons
of ammonium content and mineralogy suggestthat most of the ammonium, even in the most zeolite-richrocks, is
held by secondary K-feldspar and not by zeolites. This is confirmed by the relatively small proportion of
cation-exchangeable
ammonium in the rocks.
Keywords: ammonium, tuffs, zeolites,K-feldspar, Karlovassibasin, Samos,Greece.
SoruuernE
Les tufs miocbnesdu bassinde Karlovassi,sur l'ile de Samos(Grbce),ont largement6t6 transform6en assemblages
de silicates diag6n6tiquessecondaires.Des 6chantillons repr6sentatifsde ces roches ont 6te analys6spour I'ion
ammonium. L'ammonium est pr6sentdans toutes les rochesi zeolites(29-579 ppm,276 ppm en moyenne),mais les
teneurssont plus 6lev6esencoredans les tufs enrichis en feldspath potassiqueauthigbne(316-913ppm, 504 ppm en
moyenne). D'aprds une 6tude des teneurs en ammonium en fonction de la mindralogie, la majeure partie de
I'ammonium, mOmedansles rochesles plus riches en z6olites,se trouverait dans le feldspathpotassiqueet non dans
la fraction z6olitique. Cette hypothise concorde avec la proportion relativementlimitee d'ammonium 6changeable
dans cesroches,
(Traduit par la R6daction)
Mots-clds:ammonium, tufs, z6olites,feldspathpotassique,bassinde Karlovassi,Samos,Grdce.
INTRoDUCTION
rocks have not yet been investigated.Zeolitic tufts
arepotentially a very favorablehost for ammonium
because of the cation-exchangeproperties of
zeolites. Moreover, the natural occurrence of
ammonium in zeolitic rocks is of more than
academicinterest, becauseof the attention now
being paid to their value in agriculture and water
treatment(Pond & Mumpton 1984,Kallo & Sherry
1988).
study to find
This is essentiallya reconnaissance
out how much ammonium is present in a major
occurrenceof bedded zeolitic rocks, and to find
out if there is any relationship between their
ammonium contents and their mineralogy. The
material studied is from the Miocene Karlovassi
basin on the island of Samos, in the Aegean Sea
The ammonium ion is widely distributed as a
minor constituentofigneousrocks.It behavesvery
much like a trace element, substituting isomorphously for potassium in the rock-forming
minerals. However, freshly eruptedvolcanic rocks
and their minerals do not contain ammonium.
Although it is presentin magmas,at atmospheric
pressureit is lost by volatilization at the temperature of eruption. Any ammonium present in
volcanic and pyroclastic rocks is therefore secondary in origin.
There have beenmany studiesof ammonium in
different types of rock, including sedimentary
(Williams et al. 1989), igneous (Hall 1988) and
metamorphic (Duit et al. 1986), but pyroclastic (Fie.l).
423
424
THE CANADIAN MINERALOGIST
SAMOS
r..--:=l
Neogcne rocks
q-Pkm
Frc. l. The Neogenesedimentarybasinsof the island of Samosin the easternAegeanSea.
appreciable proportion of smectite. Carbonate
mineralsare locally present,but are not abundant.
Petrographic studies and geological mapping
During the late Miocene,the Aegeanregion was
that the main sourceof authigenicsilicates
suggest
characterizedby the development of ephemeral
glassy
pyroclastic detritus. Zeolites formed
was
shallow basins, in which either marine or nonboron-bearingK-rich feldspar formed
marine evaporiteswere deposited.The region was first, and a
and alkalinity of the pore fluids
volcanically active, and many of these shallow later, asthe salinity
is a zonal distribution of
There
increased.
basins were partly filled with volcaniclasticsedisilicates, as shown in Figure 2, with
authigenic
ments.
enrichmentin secondaryK-feldsparbeing confined
The Karlovassi basin is in the western part of to the central part of the basin.
Samosand hasan extentof approximately100km2.
During the Miocene,the Karlovassibasin was filled
METHODS OF ANALYSIS
with carbonates, mudstones, ash, cherts and
porcellanites. Subsequently,the tuffaceous hori
The samplesanalyzedin this study were selected
zons underwentextensivediageneticalteration in a
saline-alkaline lake environment, resulting in the from the four principal tuffaceous rock-types,
formation of rocks rich in zeolites,boron-bearing namely: clinoptilolite-, analcite-, chabazite-, and
K-feldspar(Stamatakis1989a)and other authigenic K-feldspar-rich members, the mineralogy having
silicates.Many of the tuffs are almost completely been determinedby X-ray diffraction. In addition
composedof authigenicsilicates,and contain only to ammonium, many of the sampleswere analyzed
minor amounts of detrital minerals. Some are for boron, by a colorimetric method, and potasvirtually monomineralic with respectto authigenic sium, by flame photometrY.
The method of analysisfor total ammonium was
silicates,especiallyclinoptilolite or boron-bearing
K-feldspar, but the majority are of mixed mineral- as follows. Each sample was decomposed by
ogy (Stamatakis 1986, 1989b). Three feldspar digestionin HF at room temperaturefor one week,
fractions can be distinguishedby X-ray diffraction: and ammonia was then separated by Kjeldahl
primary Na-rich alkali feldspar, primary distillation from alkaline solution, as describedby
plagioclase,and secondaryK-feldspar. Phyllosili- Urano (1971). The separated ammonia was
catesare presentonly in trace amounts, exceptin measuredcolorimetrically by the indophenol blue
the chabazite-bearingrocks, which contain an method, as described by Mann (1963). The
Grolocrcal
BecrcnouNo
425
AMMONIUM IN TUFFS, SAMOS, CREECE
TABLE I. AMMONIUM C1)N'IENISOF EOTXTIC TUFFSFROMSAMOS
NH4r Opn)
[email protected]
Aralcite-rich
DDtt
DD9d
DDIOb
DDloc
DDIOd
DDl0dl
SM36p
st\fi'6 !
SM36rl
SM36x
sM50 rh
407
165
215
294
m
285
326
3A
37O
208
363
anl lf,s cpt sm
anl lds dol ufs cpt
anl Ks ufs iI m
anl lf,s nafs sm
tfsulmil
anl lf,s mfs cal m il
aol lf,s m
ul Ks nals s ill
lfi anl nafs cal s
anl lf,s s
anl kfs nafs
ChabaziEiich
PB9d
PB9dl
PB9d2
l8l
282
16
cbz m pl
cbz s Dl nafs
sm cbz irafs
Clitroptilolie{ich
DDI b
DDI d
DDI el
DDI e2
DDI
DDs
sM26 I
sM:|o b
SM96a
sM96 b
SM99a
33t
412
n4
141
315
3n
113
296
192
519
29
cpt lds
spt kfs
cpt K3
6t lf,s
;t Ks naft ilt
ciltf,smin
ot lf,s
cir kfs aafs
cpr lfs
cpt kfs m
cpr nafs
K-feldsp6-iich
DDr/b
SM26n
SM33c
sM31
sMu2
SM34c
SM36o
sM99 b
sM156
sM300
413
@2
5m
387
t1l
432
483
6m
316
913
lf,s
tls
kfs
lf,s
tfs
kfs
kfs
lf,s
kfs
kfs
nafs ul sm
cpt
nafs m
nsfs
nafs dol
oafs dol
oafs sm dol
nafs cpt
ql=alcie,
cbz=chabuite,
cp!=clisoptilolite,
anl=dcite,
Abbryiatiw
dol = [email protected]€, ill = illile, kfs = K-feldspa, nafs = Na-feldsp6, pl = [email protected],
m = [email protected]
lem @d nmbqs:
[email protected] [email protected]: @h [email protected] is daignaEd by uppq.{w
urltiple emples Aom the sm sib @ dGignded by lexlq sufEs [email protected] I (uppqmi)
o z (Iowmw),
ud [email protected] sbpl6 ftom tlE s8re bed @ dFignaed by Dubq
suffre8.
Frc. 2. Geological map of the Karlovassi basin
(Stamatakis1989b),showingthe samplelocalitiesand
the zonation of authigenicsilicates.
were examined by XRD; it was found that the
zeoliteswere partly decomposedby this treatment.
Most of the clinoptilolite and more than half the
chabazite were removed, but the analcite was
relativelyunaffected.In practice,the KCI technique
proved to be less satisfactory than the others
reproducibility of the resultsis approximately + l0
becauseso much Ca2* was extractedfrom someof
ppm. The data are listed in Table l.
the samples(especiallythose containing chabazite)
A limited number of samplesalso wereanalyzed that it interferedwith the colorimetric procedureat
for exchangeable
ammonium. Three methodswere Ievelsof ammonium below l0 PPm.
tried: the first was similar to that used by soil
scientists,namely extraction of exchangeablecaAMMoNIUMConrENts oF THETUFFS
tions by 2M KCI solution for 24 hours at room
temperature.The secondinvolved extractionby lM
The range of ammonium contentsfound in the
sodium carbonatesolution for 5 minutes at l00oc,
which usuallygivessimilar resultsto KCI extraction tuffs is 29-913 parts per million, with an average
if applied to rocks containing clay minerals. The of 341ppm. Theselevelsare far abovethosefound
third required extraction by 2M HCI for 8 days at in freshly eruptedlavas, although similar or higher
room temperature.In the first method, ammonium ammonium contents are found in volcanic rocks
wasdeterminedcolorimetricallydirectly on the KCI that haveundergonehydrothermal alteration (Hall
extract; in the secondmethod, the extraction was 1989, Ridgway et al. 1990).There are significant
carried out in a Kjeldahl distillation flask, with differences between the major types of tuff
ammonium determinedon the distillate, and in the distinguishedon the basis of mineralogy (Fig. 3).
third method the extraction was carried out in a Analcite-bearingtuffs vary over a comparatively
centrifuge tube, with the ammonium determined limited range (165-407ppm), with a mean content
after distillation. The residuesof the HCI extraction of 303 ppm. Clinoptilolite-bearingtuffs are much
426
THE CANADIAN MINERALOGIST
Analcite-rich +
Clinoptilolito- rich +
aaaaa-aa
K-feldspar-rich +
0
500
NHaa(ppm)
1000
Ftc. 3. The range of ammonium content in relation to
the mineralogy of the Karlovassi tuffs.
more variable, ranging from 29 to 579 ppm, but
with a meancontentof 279 ppm, which is not much
different from that of the analcite-bearingtuffs.
The K-feldspar-rich tuffs also are highly variable
(316-913ppm), but with a considerablyhigher
averageof 504 ppm.
Thereis no geographicpattern to the distribution
of high and low ammonium values. Ammonium
contents exceed300 ppm in all the samplesfrom
the central part of the basin, whereK-feldspar-rich
rocks occur, but high valuesalso occur at the most
northerly and southerly sites (483 and 600 ppm,
respectively).
One of the most remarkablefeaturesis that large
variations occur at single sites, for example 36 to
282 ppm at PB9, 173to 602 ppm at SM26, and 29
to 600 ppm at SM99. At two of the six localities
wherethe sampleswere collectedat more than one
stratigraphic level, SM96 and SM99, respectively,
there is a strong downward enrichment in arirmonium, as follows: 192 ppm in the upper bed,
579 ppm in the lower bed at SM96, and29 ppm in
the upper bed, 600 ppm in the lower bed at SM99.
However, at the other localities, the vertical
variation is much smaller,and ammoniumincreases
upward or varies inconsistently.These variations
most likely are due to mineralogical differences
among individual samples,but the possibility of
recentleachingor supergeneenrichmentcannot be
ruled out.
Previous studies have shown that mo$t of the
ammonium in silicate rocks is accommodatedby
potassic minerals, such as feldspar and mica
(Honma & Itihara 1981, Sterne et ol. L982, Hill
1988,Hall & Neiva 1990).This is to be expected,
since ammonium has a larger ionic radius than
potassium, and cannot easily substitute for any
smallercation. Zeolites,too, are potential hostsfor
ammonium, and ammonium-substitutedzeolites
have successfullybeen s1'nthesized,although they
havenot beenrecordedin nature. In the Karlovassi
tuffs, feldsparsand the zeolite-groupminerals are
therefore the main potential hosts for the ammonium ion, but evidenceis given below that the
K-feldspar in these rocks holds more ammonium
than the zeolites.
Apart from the generally higher level of
ammonium in the feldspar-rich rocks compared
with those rich in zeolites,there is direct evidence
from the samples that are most nearly
monomineralic. Among the clinoptilolite-rich
samples, there are three that contain very little
feldspar(SM96a,SM26l, and DDle2). All of these
have lower-than-averageammonium for clinoptilolite-bearing samples(<200 ppm). One of the
clinoptilolite-rich samples(SM99a) has no secondary feldspar at all, and its ammonium content is
very low (29 ppm); it contains sodic sanidine,but
this is a primary volcanic mineral and would
therefore not be ammonium-bearing.In contrast,
five of the feldspar-richrocks contain only feldspar
or dolomite, but no zeolitesor phyllosilicates,and
these samples have ammonium contents ranging
from 371 to 913 ppm. However, the feldsparin
some of these rocks is a mixture of secondary
K-feldspar with primary sodic sanidine,so that the
actual amount of ammonium in the secondary
K-feldspar must therefore be even higher. Additional confirmation that most of the ammonium is
contained in feldspar rather than clinoptilolite
comesfrom two localities(SM26and SM99) where
a clinoptilolite-rich bed has been sampledcloseto
a feldspar-richbed; in each case,the feldspar-rich
material has a much higher level of ammonium
than the clinoptilolite-rich rock.
Among the analcite-bearing rocks, none are
nearly monomineralic.However, the analcite-bearing samplesfrom two localities (DDl0 and SM36)
include sets collectedfrom different beds at each
locality, and these show a correlation between
ammonium content and modal composition. At
each locality, the most feldspathic bed has the
highest ammonium content. The chabazite-rich
samplesare the least informative with regard to
which mineral hosts most of the ammonium;
sample PB9d2 suggestsa low ammonium content
for both smectite and chabazite, and this is
supported by the lack of cation-exchangeable
ammonium in this sample (seebelow).
The relationshipbetweenammonium and potassium in the altered tuffs is illustrated by Figure 4.
SinceK-feldspar is the main potassium-richphase
in these rocks, the rough correlation between
ammonium and potassiumis again consistentwith
K-feldsparbeing the principal host for ammonium.
The associationof ammonium with boron is a
particularly interesting feature of these diagenetically altered tuffs. Their enrichmentin boron has
previouslybeendescribedby Stamatakis(1989a,b),
and the few samplesfor which both constituents
have been determinedare listed in Table 2. Both
constituents are abundant in the zeolite-bearing
AMMONIUM IN TUFFS. SAMOS, GREECE
cllrDptlloltto- &h
Analcite- rbh
Fobspar- &h
eCL
g
'r
2
427
I
a
7oo
600
500
lata
It
.
A
100
910111213
KrO (%)
Frc. 4. The relationshipbetweenammonium and potassiumin samplesof differing
mineralogicalcomposition,
tuffs, but the highestlevelsof eachare attained in
the rocks containing secondaryK-feldspar. Martin
(1971)has previously shown by infrared spectroscopy that ammonium is not a major constituentof
boron-bearing feldsparsfrom California, but the
new data for the Samossamplesshow that it can
be a significant minor constituent.
The associationbetweenboron and ammonium
has been noticed in other altered volcanic rocks.
namely the spilites from southwesternEngland, in
which ammonium is more closely correlated with
boron than with any other trace element (Hall
1990).The associationof ammonium with boron
also is of interestin that the ammonium and borate
ions would both have contributed to the alkalinity
of the solutions involved in diageneticalteration.
TABLE 2. COMPARISON BETWEEN THE BORON AND AMMONII,JM
CONTENII OF ALTERED TUFFS
K2o
(%')
Minemlogy
B3r
Gpn)
NIL+
(ppm)
245
192
2.51
Clinoptilolite-rich
410
363
7.50
Analcite-dch
SM26n
84
g2
9.23
K-feldspu-rich
sMl4
910
3W
10.95
1341
316
9.41
SM96a
sM5oth
sMl56
Ftxso eNIoExcHeNcsesLEAMMoNIUM
Representativesampleswere selectedfor measurements of exchangeableammonium by three
different methods, and the results are given in
Table 3. In none of the samplesis more than a
small proportion of the ammoniumreadily exchangeable,evenin thoserocks in which a zeoliteis the
main potentially ammonium-bearingmineral.
Stevenson(1962) found that in a selectionof
(mostly weathered)igneous rocks, an averageof
20a/oof the ammonium could be extracted by
treatment with cold lM KCI for 2 hours. This is a
measure of the ease with which ammonium is
releasedfrom secondaryclay minerals by cation
exchange. In contrast, less than l0 ppm of
ammonium was extractedfrom most of the rocks
in Table 3 by a 24-hour treatment with cold 2M
KCI.
It appears that although cation exchangeis a
characteristicproperty of the zeolite group, much
longer times are neededfor ammonium exchange
in zeolites than is the case for clay minerals.
Individual zeolitesvary not only in their cation-exchange capacity, but also in their selectivity for
particular cations,dependingon their structureand
in particular the size of channelsthrough which
cations can migrate. Information on the exchangeability of cations in natural zeolites is very
incomplete, but as an exampleof the differences,
clinoptilolite is hiehly selectivefor Cs over all other
428
THE CANADIAN MINERALOCIST
TABII 3. CATIOI{iEXCHANGEMEASUREMENTSON SELECTM SAMPLES
,
Smpls
Minqalos/
DDdr
Aralcie-rich
DDgd
ToalNlt4+
(ppn)
PB9 d
<10
l9
165
<10
1)
<10
l6
29
JOJ
Oralazit€{ich
PB9 (D
DDld
NHa+ (ppd)
q7
DDIOd
SMsoth
Erchegable
181
<10
36
<10
t4
4r2
<10
315
<10
17
"
296
13
30
DIt/b
K-feldspr-rich
4t3
<10
A
SMlte
"
432
23
37
ClinoptilotiE-tich
DD'
sl{:nb
22
39
SM99a'?9518
Cffimrchqgrtrcatmffts:
A 2MKCI(24hoBar2fq):
minutosat 100qC);C 2M HCI (8 daysa! [email protected])
6E
B lM Na2C03(5
cations,whereasanalcitewill not easilytake up Cs
in exchangefor Na (Ames 1960, Barrer 1982).
Similar variationspresumablyexistin regardto the
ammonium ion. However. the HCI treatment
applied to thesesamplesnot only involved cation
exchangeof H+ for NH4+, but actually led to
partial dissolution of the zeolites(almost complete
in the case of clinoptilolite), so that the small
amount of ammonium extractedmay be taken as
a further confirmation that only a small proportion
of the ammonium in these rocks is held bv the
zeolites.
Dtscusstott
The possiblesourcesof ammoniumin tuffaceous
rocks such as those in Samosare as follows: (l)
primary ammonium of magmatic origin, or (2)
secondaryammonium, introduced during deposition, diagenesisor weathering.
The amount of ammonium that may have been
presentin the magma originally cannot be known.
On the basis of the amount of ammonium iu
granites, which is typically in the order of 10-100
ppm (Urano 1971,Hall 1988),the rhyolitic parent
magmas of the Samos tuffs would undoubtedly
have been ammonium-bearing,but all ammonium
compoundsare unstableat magmatictemperatures
under the pressureand ammonium fugacity of the
atmosphere,and all magmatic ammonium would
have beenlost on eruption. This point needsto be
emphasized,becausethere are publishedresultsof
analysesshowing more than l0 ppm ammoniacal
nitrogen in some volcanic rocks (compilation of
Wotzka 1972). Such data need to be questioned,
because(a) the existing analytical data commonly
show chemicallycombinednitrogen (i.e., including
the ammonium ion) exceedingtotal nitrogen (as
measuredby vacuum degassing),which cannot be
correct, and (b) most existing analytical data do
not distinguish between fixed and exchangeable
ammonium, the latter being presentin significant
amounts in almost.all rocks that have undergone
even a small degree of surface weathering.
Unpublished data by one of the authors (AH) on
sevenvery fresh samplesof volcanic rocks ranging
from rhyolite to basaltshowedfixed ammonium ro
be below the limit of detection ( - 3 ppm) in every
sample;on that basis,freshvolcanicrocks probably
do not contain ammonium. All the ammonium
found in the Samos rocks is thus taken to be of
secondaryorigin.
There are many possible sourcesof secondary
ammonium. A small amount of ammonium is
presentin all surfacewaters and sediments,and is
a product of biological activity. Ammonium could
have been introduced into the tuffs at the time of
depositionfrom the lake waters,from sedimentsin
the lake, or by addition of water to the lake,
especiallyif fed by hot springstapping ammonium
from rocks of the surrounding area. Ammonium
could have beenincorporatedduring the recrystallization that gave rise to the assemblagesof
authigenic minerals, or by interaction with
groundwater up to the presentday.
Whatever the source of the ammonium, its
incorporation in the rocks is governedby its large
ionic size. Becausepotassium is the only major
elementfor which it can easilysubstitute,the only
rock-forming minerals that carry ammonium are
those that also can contain potassium, i.e.,
feldspars, micas and zeolites. Micas are not
significant constituentsof the Samostuffs, and the
distribution of ammonium in these rocks was
therefore determined by the crystallization of
K-feldspar and zeolites. These minerals differ
greatly in their ammonium-bearing capabilities.
The ammonium ion can be taken up by zeolitesby
cation exchangefor Na, K or Ca, but is just as
easily displacedby those elements,so that during
a long period of geological time, the ammonium
contentof the zeolitescan riseand fall with changes
in the composition of ambient hydrothermal
solutions or groundwater. In contrast, any ammonium entering K-feldspar is fixed, and is not
subject to removal by cation exchangeunlessthe
feldspar is replacedby another mineral.
AMMONIUM IN TUFFS. SAMOS. GREECE
The time scale of diageneticrecrystallizationin
the Karlovassi tuffs is still largely unknown, but
on the basisofthe aboveconsiderations,
a sequence
of events that could account for the present
distribution of ammonium is:
(1) Deposition of ammonium-freevolcanic detritus
in a lake basin.
(2) Early diageneticrecrystallizationof fine-grained
volcanic glassy material to give zeolites, by
saline-alkalinesolutions containing ammonium.
(3) Later diageneticconversionof ammonium-bearing zeolites to ammonium- and boron-rich Kfeldspar by highly saline-alkalinesolutions.
(4) Partial removal of ammonium from the zeolites
surviving from stage (3) by groundwater low in
ammonium but higher in other dissolvedcations.
This sequenceshould be regarded only as a
simple working hypothesisuntil more is known of
the timing of diagenetic events. Whereas the
K-feldsparprobably crystallizedduring a restricted
time interval when diageneticwaters were at their
hottest, the zeoliteswere subjectnot only to cation
exchange,but also to continuing recrystallization
over a much longer period, perhaps extendingup
to the presentday.
ACKNowLEDGEMENTS
The authors thank Professor R.F. Martin for
checkingthe identity of the feldsparphasesin some
of our feldspar-richsamples,and Dr. J.N. Walsh
for determiningthe K2O contentsgiven in Table 2.
429
& Nerve,A.M.R. (1990):Distributionof the
ammonium ion in pegmatites, aplites and thetr
minerals from central northern Portugal, Mineral,
Mog. 54,455-461.
HoNrua, H. & Irrrnna, Y. (1981): Distribution of
ammonium in minerals of metamorphic and
granitic rocks. Geochim. Cosmochim, Acta 45,
983-988.
Kar-r-o, D. & Surnnv, H.S. (1988): Occurrence,
Properties and Utilization of Natural Zeolites.
Akademiai Kiado, Budapest, Hungary.
MerlN, L.T. (l 963): Spectrophotometric determination
of nitrogen in total micro-Kjeldahl digests:application of phenol-hypochlorite reaction to microgram
amounts of ammonia in total digest of biological
material. Anal. Chem. 35, 2179-2182.
Manrrn, R.F. (1971):Disorderedauthigenicfeldspars
of the series KAlSi3Os-KBSi3O6 from southern
California. Am. Mineral. 56, 281-291.
PoNo, W.G. & MuvproN, F.A. (1984): Zeo-agriculture: Uses of Naturdl Zeolites in Agriculture and
Aquaculture. Westview Press, Boulder, Colorado.
Rrocwav, J., Apprsror, J.D. & LrvtNsoN, A.A.
(1990): Ammonium geochemistry in mineral exploration - a comparison of results from the
American Cordilleras and the southwest Pacific.
Appl. Geochem. 5, 475-489.
Srauerarrs, M.G. (1986): Boron Distribution in Hot
Springs, Volcanic Emonations, Marine Evaporites
and Volcanic and Sedimentary Rocks of Cenozoic
Age in Greece.Ph.D. thesis, Univ. Athens, Athens,
Greece.
REFERENCES
Aurs, L.L. (1960): The cation sieve properties of
clinoptilolite. Am. Mineral. 45, 689-700.
Bennrn, R.M. (1982): Hydrothermal Chemistry of
Zeolites, Academic Press, London.
Durr, W., JeNseN,J.B.H., VeN BnrsvrN, A. & Bos,
A. (1986):Ammonium micasin metamorphicrocks
as exemplified by Dome de I'Agout (France). Am.
J. Sci. 2E6,702-732.
Hall, A. (1988): The distribution of ammonium in
granites from south-west England. J. Geol. Soc.
London 145. 37-41.
(1989): Ammonium in spilitized basalts of
southwest England and its implications for the
recycling of nitrogen. Geochem. J. 23, 19-23.
(1990): Geochemistry of spilites from southwest England: a statistical approach. Mineral.
Petrol.41, 185-197.
(1989a): A boron-bearing potassium feldspar
in volcanic ash and tuffaceous rocks from Miocene
lake deposits, Samos Island, Greece.Am. Mineral.
74.230-235.
(1989b): Authigenic silicates and silica
polymorphs in the Miocene saline-alkaline deposits
of the Karlovassi Basin, Samos, Greece. Econ.
GeoL84.788-798.
SrsnNr, E.J., RrvNor-os,R.C. & ZaNtor, H, (1982):
Natural ammonium illites from black shales
hosting a stratiform base metal deposit, Delong
Mountains, northern Alaska. Clays Clay Minerals
30, 16l-166.
SrEvENsoN,
F.J. (1962):Chemicalstateofthe nitrogen
in rocks. Geochim. Cosmochim. Acto Xi, 797-809.
UnaNo,H. (1971):Geochemicaland petrologicalstudy
of the origins of metamorphic rocks and granitic
rocks by determination of fixed ammoniacal
nitrogen. J. Earth Sci., Nagoya Univ. 19, l-24.
430
THE CANADIANMINERALOGIST
Wrlnnvs, L.B., FEnner-1,R.E., CnrrN, E.W. &
SessEN,R. (1989): Fixed-ammonium in clays
associated with crude oils, Appl. Geochem, 4,
605-616.
WLorzra,
F. (1972): Nitrogen. /n Handbook of
Geochemistry (K.H. Wedepohl, ed.), vol. II,
Chapter 7B-M,O. Springer-Verlag,Berlin.
Received March 5, 1991, revised manuscript occepted
September3, 1991.

Similar documents