THE AMERICAN MINERALOGIST. VOL. 48, NOVEMBER-DECEMBER, 1963

A STUDY OF ALLUVIAL MONAZITE FROM MALAYA1

B. H. Fr,rNrax, Geologi.cal Survey, Fed'eration oJ Molaya; l. R-

Burr-nn, Department of Geology, Imperial College, Lond.on; l^No

G. M. HennAL, Oi)erseas Geological Surveys, LoniJon.

Ansrnecr

sixteen samples of alluvial monazite from Malaya have been analyzed for their rare

earth, thorium, uranium, calcium, and silicon contents, and an attempt is made to find

some correlation between composition and specific gravity, mass magnetic susceptibility

and interplanar spacing. Compositionally, three of the samples, all {rom columbite-

producing areas, are unusual, one being especially rich in Eu, one in Gd, and one in U

and Dy.

INrnonucrroN

Monazite is of widespread occurrence in the alluvial deposits of

Malaya and is a major by-product of tin-mining. This is the first time the

rare earth distribution of Malayan morrazite has been examined in de-

tail, and the study was made as an init ial step towards the evaluation of

the mineral's commercial possibil i t ies. Emphasis was thus placed on the

alluvial deposits along the west coast of the country where mining oper-

ations are concentrated. The geographic distribution of the samples

examined is shown in Table 5.Determination of the individual rare earths was done at the Imperial

College of Science & Technology, London, by J. R. Butler, whereas all

the chemical assays were carried out at the geochemical laboratories of

the Geological Survey of Malaya under the supervision of G. M. Harral.

PnBpanarroN oF THE Mn:rBnrar

In order to enhance the practical value of this study, the samples used

were made to approximate as closely as possible to the material that

could be extracted commercially. From experience of Malayan alluvial

monazite, commercial extraction can be duplicated by using a Frantz Iso-

dynamic electromagnetic separator with a setting of 25o for the side slope

and 15o for the forward slope, and extracting the material within the

range 0.8 to 1.0 amps., i.e. all material magnetic at 0.7 amps and non-

magnetic at 1.0 amps was discarded. This also removed any xenotime

which is more magnetic. At least 99.6/6 of the monazite was extracted

within the range 0.8-1.0 amps for all the samples except Nos. 2, 8, 9, 10,

and 15. fn samples 2, g, !0, and 15, 0.87a, 3.57a, 2.07o, and 8.6/6 re'

spectively were attracted at 0.7 amps, while in samples 8 and 10, 2.1/6

and 2.3/6 respectively were not attracted at 1.0 amps.The retained fraction was then cleaned by hand-picking under a stereo-

1 Government copyright is reserved.

t2l0

MALAYAN ALLUVIAL MONAZITE

microscope, all material containing inclusions being discarded. Thecleaned sample was then split in two, half being used for chemical analy-sis and half being used for the other tests.

GBNrner DBscnrprrow oF THE SauprBs

As all the samples studied are detrital, their derivation cannot becategorically stated. It is almost certain that most, if not all, are derivedfrom cassiterite-bearing granite, the major rock type in Malaya. probableexceptions are samples No. 1a, No. 15 and possibly No. 1. These may bederived from pegmatites which occur within the two localities in closerproximity than any granite mass. However, in appearance, No. 15 doesnot resemble known pegmatitic mona"zite from the same area, and Nos. 1and 1a look even less l ike it.

Apart from No. 14 which was obtained from the heavy mineral suite ofa beach sand, all the samples come from areas which are being activelymined for cassiterite. Nos. 1 and 1a were obtained from a single alluvialconcentrate and, with No. 15, occur in columbite-producing areas.

In form, all the samples were fairly similar, being moderately well-rounded discrete crystals or parts of crystals. No. 1a was appreciably lessrounded and had better developed crystal faces than the others. No. 14was the most irregular, rnuch of the material being fairly angular. Nos. 1and 15 were slightly more angular than the remainder.

No. 14 was also the most variable in size, the material ranging from0.1 to 0.4 mm. No. 10 had a similar size range but was mainly 0.3 mm.The remainder ranged from 0.2 to 0.6 mm; samples I,La,3,6, and 12 wereslightly coarser than the rest.

In color, all except No. 1a were similar, varying from a clear, deepcanary yellow for fresh material, through cloudy-white translucent ma-terial, to completely opaque, cream-colored material with a resinousluster as the rare earth oxides became progressively more oxidized. Inbulk the altered material imparted a pale greenish tinge. In a few casesonly, oxidation gave rise to a dark gray or black coloration. Nos. 1 and 9were almost completely fresh, being a clear deep yellow, the former hav-ing a pale salmon tinge. Nos.3,8, and 11 were completely transformed toopaque, cream-colored material. The remainder had varying proportionsof oxidation. No. 1a was strikingly distinctive in color, being an intense,clear apple-green. Green monazite is known in quantity only from theSemeling area in Kedah, although traces of a much paler green monazitewere noted in samples 5 and 10.

The associated minerals are:

Nos. 1 & la. Columbite, cassiterite, garnet (mainly pink to orange almandine, very minorpale yellow spessartite), tourmaline, gahnite, magnetite and minor amounts

t212

No. 2.

No. 3.

B. H. FLINTER, r. R. BIITLER AND G. M. HARRAL

of wolframite (?), Ta/Nb-rutile, chromite and fergusonite' In the separated

Iraction of No. 1, inclusions of gahnite were fairly common but columbite

was less abundant. In No. 1a, inclusions were abundant, the majority being

columbite which was irregular in shape and randomly distributed. A few

crystals of tourmaline were also noted as inclusions. A most interesting

feature was the fairly common intergrowth of the monazite with crystals of

a beige-colored, opaque, resinous variety of zircon. A few typical examples

are shown in Fig. 1.Cassiterite, zircon, ilmenite, iron oxides, xenotime, Ta/Nb-rutile, tourmaline,

columbite, and minor amounts of allanite and euxenite. The clear monazite

was free of inclusions. The translucent resinous (oxidized) monazite con-

l m m

Fro. 1. Intergrowths of monazite (dark) and zircon in sample la.

tained a fair amount of inclusions, these being cassiterite, Ta/Nb-rutile and

columbite.Cassiterite, zircon, ilmenite, iron oxides, tourmaline, xenotime, Ta/Nb-

rutile and a minor amount of columbite. Inclusions of Ta/Nb-rutile were

present.

No. 4. Similar to No. 3, but with epidote and minor allanite additionally present,

and Ta/Nb-rutile very abundant. Inclusions of cassiterite and Ta/Nb-rutile

were rare.No. 5. Similar to No. 3 with some'anatase additionally present. An appreciable

amount of columbite occurred.No. 6. Mainly ilmenite and Ta/Nb-rutile, with less zircon and cassiterite. Some in-

clusions of Ta/Nb-rutile were present.No. 7. Xenotime, ilmenite, cassiterite, and zircon. Some grains contained inclusions

of ilmenite.No. 8. Similar to No. 3 except for the absence of columbite'No. 9. Similar to No. 3. Inclusions were very rare.No. 10. Similar to No. 9.No. 11. Similar to No. 3, except that there were fewer inclusions.No. 12. Similar to No. 3. Inclusions were absent.

MALAYAN ALLUVIAL MONAZITE L213

No. 13. Similar to No. 3 with a minor amount of gahnite additionally present. fnclu-sions of Ta/Nb-rutile occurred in minor amounts.

No. 14. Mainly xenotime and ilmenite, with minor amounts of tourmaline, columbite,allanite, Ta/Nb-rutile, cassiterite, and zircon. The separated fraction wasvery pure, only very minor amounts of finely disseminated opaque inclusionsoccurring.

No. 15. Columbite, cassiterite, Ta/Nb-rutile, gahnite, garnet (pink almandine), zir-con, pyrite, and tanteuxenite. In the magnetically separated fraction, inclu-clusions of Ta/Nb-rutile and cassiterite were fairly abundant.

ANeryrrc,lr MBrgoos

Total rare earths, SiOz and CaO were determined gravimetrically;so was thOz (by the double hexamine method of Ismail and Harwood,1937). Ce was determined titometrically on a solution of the rare earthoxides after oxidizing Ce+ to Cea+ with ammonium persulphate. U wasdetermined colorimetrically after the method of Guest and Zimmerman(1955). Individual lanthanons were determined spectrographically usingthe chemically separated rare earth oxides (Butler, 1957 a). Possible inter-ference due to the coincidence of Ca4435.68 A with Eu4435.53 A *utruled out by confirming the absence of the stronger Ca line at 4434.96 L.Y was determined in the rare earth oxides by an r-ray fluorescent methodwith added Sr as an internal standard to Y.

Repeat specific gravity determinations were made on a Berman bal-ance using the powder basket method with CCI+; and the results are+0 .01 .

DrsrnrsurroN oF TIIE R'lnB Eenrns

Rare earth distribution has been studied more in monazite than in anyother rare earth mineral and the general abundance pattern is now fairlywell known. Results have been correlated with the work of Carron et al,.(1958) who found that laboratory orthophosphates of La, Ce, Pr, Nd,Sm and Eu each had a rnonazite type structure whereas those of Ho, Er,Tm, Yb, Lu and Y had a xenotime type structure; (Gd, Tb) POa (withGd:Tb about unity) crystall ized as the monazite phase but with minoramounts of the xenotime phasel (Tb, Dy) POa (with Tb: Dy about unity)crystallized as the xenotime phase but with minor amounts of the mona-zite phase. High contents of Y and the heavy lanthanons are not, there-fore, to be expected in monazites.

There is much natural variation in the relative abundances of Y andthe l ight lanthanons in monazite even though the vast majority of thoseanalyzed are from granites, granite pegmatites or alluvial deposits de-rived therefrom. Wylie (1950) Vainshtein et al. (1955, t956a,1956b) andSeminov and Barinskii (1958) give data for the light lanthanons and

I2I4 B. H. FLINTER, J. R. BUTLER AND G. M. HARRAL

Murata et al. (1953,1957, 1958) give data for Y and the l ight lanthanonsdown to Gd. Recently, Heinrich et al. (1960) have studied 14 pegmatiticmonazites for La, Ce, Pr, Nd, Sm, Gd and Y and show that: (i) para-genetically identical monazites from a single pegmatite can be remark-ably similar in their rare earth composition (ii) paragenetically identicalmonazites from difierent pegmatities from the same districts may havecharacteristic (regional) rare earth distributions.

fn order to see how the data for the Malayan monazites ('Iables 1 and2) compare with those for other monazites, it is convenient to considerthe range and variation in concentrations of pairs of lanthanons. Table 3shows ratios fior Ce/La, Ce/Nd, La/Nd, and Nd/Sm from the varioussources indicated. Ionic radii for La3+, Qsa+, NTds+, and Sm3+ are, (Ahrenslg52), t. l4 A, t.O7 A, l.O+ A, and 1.00 A respectively. It might be ex-pected, therefore, that more variation would be obtained for the ratioLa/Nd (with a difierence in ionic radii of about l0/) than for the ratioCe/La, Ce/Nd or Nd/Sm (with differences in ionic radii of about 7, 3,and 4/6 respectively) . fn monazites derived from granites or granite peg-matites, the 12-fold variation in La/Nd is conspicuously higher than the3f-fold variation in Ce/La and the 4]-fold variation in Ce/Nd. Variationin Nd/Sm is, however, at least 7-fold. Sm varies from about 14 to about 1at. /6 in the total rare earths in monazites, a much larger variation thanlor La, Ce or Nd, and it is probably not as suitable a size for the rnonazitestructure as La, Ce, and Nd for the conditions under which monazitecrystallizes from pegmatities and granites. La, Ce, and Nd are collec-tively more essential to monazites than Sm, and it may be supposed thatSm variations in monazite reflect more closely the variations in con-centrations in Sm3+ in the fluid from which the monazites crystallize thando variations in La, Ce and Nd. (The low variation in the Pr content inthe rare earths of monazites established by Murata et al,. (1953, 1957)would indicate that the size of. Pr3+-1.06 A-is eminent for monazitescrystallized from acid rocks). Sm can, of course, exist as Sm3+ or Sm2+and the larger Sm2+ (ionic radius about 1.2 A) would certainly not be in-corporated in the monazite structure.

It will be seen from Table 3 that the Malayan monazites have averagevalues lor CefLa, Ce/Nd, and Nd/Sm very close to those obtained forpegmatitic and granitic monazites by other authors. None of the Malayansamples appioach, compositionally, the most La-enriched monazite(from Kazakhstan) recorded by Vainshtein (1956b). On the other hand,the Malayan monazites are distinct from the three cited monazitesoccurring in association with carbonate, bastnaesite and pyrite (Table3). The monazite from the pyrite-apatite vein (Murata et al., 1957) has,for instance, equal atomic proportions of La and Ce (Ross el ol., 1958),

MALAYAN ALLUVIAL MONAZITE

€

d n I i F F l N d N # d d + r F i F4 I = l I d + $ r @ : i @ N N c O

l , N - i H N $

,4R itd iF Htn dtN itd itR dft ilR ilN iild

€ a o i + 6 o N N \ o q @ \ o c o N o4 r + N 4 r n € o o o 4 \ o e . + 4 F

i ld i lR i lN dld i lN i lR iF 4R dN\O \O 4 € dt d N 0o c) 4 4 O D c|, : \Oo o \ h N + 1 N < i F o o N \ o N @

dl,5 1alo l

+ o|' \O O cO <1 N \o ts .O \O N $ O\ r- Id d H : C O F : H i Q O N

; ; - ; ; ; ; d ; - ; ; ; ; ; ; ; ;

o < 1 < 1 $ O + N + O € N + N € N €€ \ o o + r + o i o € d N D o € H o

o d J J . . . ; J . i J . i J J - ; J J J c ;

N D N O N b D O F 4 O N T N C ) 4c O b O \ N N \ O o r € N 4 O 9 O \ O N

O - O F i O O O O - O - F C ; - -

O \ c O O € 4 N H O o O i r N N H \ O Oi + + 1 N O S O N O O N T 9 . $ N 4;,- i . - : ; - : ; ^ - l - : , { - l ; ; . { - i ;

O O\ N O\ O oO <r n, cO i.O N N O : cNo c ) \ o o o a N + r e n o . + c ) o o , $.i G ro rd * <j .o <, <, * +'d

": tci + to

o r Q N - s o \ o \ b N N a o ' + c oo O o N a N o + < 1 + O N N o \ O a

J i J J o J - ; j i o - ; - J J J - ;

. t 9 1 n a ? a n n a q i o a n aN O 4 N N N N N N N N N E N N $: i H i : H i i : i d d

\ \ 9 \ 9 9 ? n 1 9 \ \ 9 \ 1 9N N N N N N O d N N N N N N N N

p

FiF

o

N

D i :

A I o r N * o o o o N + r $ < , " o N o H-S

I <. oi J.i j 6; d <, <. -; $ -; ; -i + oit f l -

+ i * H : i H

A @ * - * --.1 | G c; ; N d ; * o o <i

"d * <i d

"d ;

U ) I N N N N N N N N N N N N N N N N:k I * o\ o. n Nn h + <, @ i n o\ o\ e N-,j

I G ci ;.o G ; 6 di d ; * 6 ; ; "d

*;O I N N N N N N N N N N N N N N N N

.+ cO + \O r O O N + N € + cO : :f HO O + < i T T € + N € N T N O \ O S

...J ̂ ; J J J J J J J J J J J J J.i

O l o . r n o 6 d \ o d + o N Nl : I O N d N O D H i F O d i € - - N

c i d I o d c o . o d N N N N N N N N N N +

d H N O dr @ \or oo o\ o i N o d{ 4i : i d e i

oi

X

NQ

U

o

Q

a

n

otsrl

z

tdItr

F

o

Eza

l 2 l6 B. H. FLINTER. J. R. BUTLER AND G. M. HARRAL

I€ lI

NQ

-i

o 5. a u

o €a \ <o i

o o

a i t re H ' !

6 . N

a z - ^' ^ ^ q

.: .F ci

'": ql * o * - *r o li l D 6 + 1 o \ O *

l * - N

. . l o - d , * ^I o . . j < . o . i "

N d l

O r o < 1 O < r * O d 'N $ N

d i + d r n O o \ s: - i

" . j + , o ' < , " - j * ,N d l

< ' o + r ^ o \ N

N + + O ' \ - 4 " i < i

a qs o d r < r ^9 l ; + + o i "

l - * N

\ O o : i l O ^ N N O

N V N

o \ o d r o : r o \- j ; + o . ; * d s iN S N

c O \ O < i r c } \ ^ N O

N r t + o \ - o * O o

l o - - o ^ N D o o' o l " i + < . o ; o J - jl - s N

o \ < t l d ' o o o \ N- j . d + o < . * - i < ,N $ N

Il \ o 4 + N ^ 4 d {

- l O n d ' O - o " i <l - $

N

l a - * * o o \ r . )q l o ; . + c i - i n J o . il - *

N

@ N < r N * O \ Dd o < . ; . c ; " J " ;d s N

. . l . =ae i qqj l o , n 4 o n * : Q

l - = N

j . 4 N l t l c o ^ o . 4 n'

s + < i O \ ! O F Nl N < .

i

@H

F

?il.

rl

FroFz

F2

?.r'.

q

g

z

O

H

N

rl

F

s3f ,28a3>

1217MALAYAN ALLUVIAL MONAZITE

€

6': ? ^3E^

. v ) €

E e 3: * i r -

. E * * :t s d rH E s ?#>d

N!

N T^ ^ ) ? ^ G ' y ?n 4 = 4 \ o : . . ;g E;5 E Ja€'. v J 4 J J D A 6

, . O \ , O O F _B E . v B

t s R * \ r * * t- ' - t

g g : 3 . 9 : : 4

2 9 r r 9 9 s d s d !

ssSss>>fi

oO

ooo

N ^

O N

€q a 1o N < {

9 -g = =

- E 4

i r YY ! 4

i o

c O N O I H Q c O i c O

6 d ; ; + < , d ;

N

N

sN

< ' c ; o$ o N - 3 3 - E *. i J . i J - i c ; J c ; J

w G a6 c t \ o l 2 E q {' ' : i l qq

] =

h

9 1 1 9 - : : 1 0 :N N N N N N N N N

\ o - Es N \ o . + 6 q r

' l t li c r .+ ' 9 .h q

- J

Ea

z

zF-l

i

z(,

cdF-l

f )

c € d XO O 0 r

\ ' l < i o i i < lN d H O H

HO o o o o o. . € € € € €b o ! ! ! a qd 6 6 6 6 6

5 Fp Fg ,u.o F.o -a.o *o

o o o o o o o o d

d d d d d d d d =| ! L H E t i ! F L Jo o ( J ( , ( J ( , ( , ( , 2 i

a

zd d E. , N

N d

t r _ 9

< ao . . .

E

sN <1 =fl

. O c O On n l q q n : . :i i N N N N N O N

1218 B. II. FLINTER, ,T. R. BUTLER AND G. M. HARRAL

and the ratios Ce/Nd and La/Nd are conspicuously higher than arenormal for monazites from acid rocks. Borovsky and Gerasimovsky(1945) do record Ce/La at 0.91 for one out of 7 monazites from granites;however, the samples are from widespread localities with Ce/La aver-aging as low as 1.3, and it is believed that La has been systematicallyunderestimated. Murata et al.have indicated a compositional criterion inconsidering monazites of different parageneses. They calculated the sumof the atomic percentages of La, Ce, and Pr in the total rare earths ofmonazites and found values between 58 and 80 at. /6 for monazites asso-ciated with acid igneous rocks or acid pegmatites. The ratio Ce/(Nd+Y)(atomic) was shown to decrease as the La* Ce*Pr sum increased. Hein-rich et aI. (1960) found values down to 5l at. /6 for pegmatitic monazites.In the two monazites associated with alkali rocks, Laf Ce*Pr was 87 at.

/6.Most of the Russian data are not accurately amenable to this treat-ment as Y is seldom reported, but in mona"zite from alkali granite peg-matite, Tommat, Yakutiya, U.S.S.R. (Seminov and Barinskii, 1958),La*Ce*Pr is 81 at. /6intotallanthanons, and any Y would reduce thisvalue on a total rare earth basis.

In the monazite selected for La enrichment (from Kazakhstan) La*Ce*Pr amounts to 90.1/6 (presumably atomic /) and Y would haveto be 13 at./ein the total rare earths to bring the value for La-|CetPrdown to 80 at./6.In the lanthanons of the La-enriched monazite, Dy,Er and Yb are not reported and are presumably below 0.2/6. It seemscertain that Y would be appreciably less than 9 at. /o which is the highestvalue for Y recorded in monazites to date. The use of the sum of La* Cef Pr, therefore, has limitations, and more data on monazites from quartz-free igneous rocks or alkali rocks are needed before a paragenesis outsideacid rocks can be proved from rare earth contents and distribution alone.

Most of the Malayan monazites have the sum of the atomic percent-ages of La, Ce, and Pr in total rare earths between 69 and 73/6;theirCe/(Nd*Y) ratios vary between 1.73 and 2.05. The exceptions are Nos.la, 2, and 15 which have La* Ce* pr values of 65.4, 63.4, and 58.8 andCe/(Nd*Y) ratios of 1.56, 1.40, and 1.34. These three monazites may,theiefore, be of pegmatitic origin.

The Y contents of the Malayan monazites are comparable with thosereported by other authors. They range from 2.5 to 5.9 at. /6 and average3.8 at. /6 compared. with the range of less than 0.3 to 9.1 at. /6 and aver-age of 3.9 at. /6 found by Murata et al,. (1953,1957). However, when thecontents of Eu and the heavy lanthanons, Gd and Dy, are considered,then three of the Malayan monazites stand out as unusual.

In the Malayan samples monazite No. 1 contains easily the highestconcentration of Eu; it is nearly I at. /6 in the total lanthanons. Eu is

MALAYAN ALLUVIAL MONAZITE I2I9

rarely recorded in the literature. Borovsky and Gerasimovsky (1945)found up to 1.5/6 Eu in the (chemically separated) yttrium group of fourmonazitesl this might be equivalent to as much as 0.2/6Euin the totalrare earths and is comparable to the amount found in the Malayan sam-ples No. 6 and No. 9. Semenov and Barinskii (1953) found 0.1/6 Eu inlanthanons of two monazites but Murata et al. (1957) did not record iteven in their most Nd-rich monazite; Vainshtein et al. ma"ke no referenceto Eu in any of the many monazites that they have examined, and Hein-rich et al. (1960) did not report on Eu for their Sm-rich or other mona-zites. Not much is known about the geochemistry of Eu. It difiers fromall the other rare earths, except Sm and Yb, in being able to assume thedivalent state and this has been advanced as an explanation for itsscarcity in rare earth assemblages. Euz+ has a size rather larger than 1.2 Aand might, therefore, be expected to be camouflaged in Sr2+, Pb2+ or evenBa2+ minerals. Low concentrations of Eu have, in fact, been found in Srminerals l ike strontianite, but less Eu (below 0.05%) has been found instrontianite from Kangankunde, Nyasaland, than in the associatedmona"zite (with approximately 0.2-0.370 EuzOs in the total rare earthoxides). Eu does not seem to have been seriously sought in minerals l ikeanglesite, cerussite, barytes or witherite; nevertheless, it is fair to observethat Eu must be one of the rarest elements that is entirely lithophile andnon-radioactive. Monazite No. 1, therefore, is a very Eu-rich mineral,and in the manufacture of the metal or compounds of Eu, it would be amost valuable raw material. Many monazites from which Eu is com-mercially extracted have Eu running at only I or 2 per cent of the Sm con-tent (J. Clinch, pers. comm.).

Monazite No. 15 contains abnormally high Gd-5.1 at. /6-almosttwice as high as in the other Malayan samples. Murata et al. (1953,1957)recorded 4 at. 7o in a pegmatitic monazite with just over 8 at. /6 Y (inthe rare earths); they show others with up to 9 at./6Y butless than 4 at./6 Gd and there is no correlation between Gd and Y. Heinrich et al. (1960)record 4.9 at. % Gd and 4.4 at. /6 Y in their Sm-rich monazite. Mostmonazites for which Gd and Y have been determined show Y exceedingGd but the ratio Y/Gd can be as low as 0.5. Monazite No. 15 has thehighest of both Y and Gd of the Malayan minerals but these elements donot vary sympathetically. The reason for high Gd in No. 15 is probablyits availability during crystallization of the monazite, favorable tem-perature and rate of crystallization rather than major element composi-tional controlling factors as there is nothing exceptional about the over-all composition (Table 1). RErOs and ThOz have values paralleled byseveral of the other Malayan monazites in which Gd is quite low. CaO isfairly high, but it is just as high in No. 9 and No. 12 where Gd is average

I22O B. H. FLINTER, J. R. BUTLER AND G. M. HARRAL

or below average; neither does Ca vary with Gd or Sm more sensibly' SiOris quite low in No. 15 and also in the unusual monazite No. 1a;but SiOzis also low in No. 1 which shows no startling properties amongst the

lanthanons.Monazite No. 1a is unusual in containing approximately t.3 at. /6

Dy (and just detectable Tb) in the total rare earths; in all the other

Malayan monazites it is below 0.8 at. /6. However, Dy is not often de-

termined in monazites, and 1.3 at. /6 need not be regarded as an extreme

limit of Dy content in the rare earth oxides of monazite. Vainshtein el ol'

(1956a) detect Dy in 13 out of 56 monazites, but they give no figures, and

their l imit of detection, although not stated, would seem to be below 1/6'

DyPOn crystallizes in the xenotime class, but presence of xenotime inter-

mixed with monazite No. 1a is ruled out as a cause of the high Dy con-

tent as Yb is below about 0.1/6 in the separated rare earth oxides.

Xenotime contamination is also unlikely in view of the Iow Y content in

the monazite and the absence of any lines in the x-ray patterns. The

major element constitution of this Dy-enriched monazite is interestingin its U content. U is no less than about 6/6. This is at the expense of a re-

duction in total RE20a and not a reduction in ThOz. (Indeed ThOz is also

abofi 6/ .) fn rare earth minerals where substantial U occupies rare

earth lattice positions, the heavy lanthanons tend to preponderate' Thus

members of the euxenite-polycrase series and the samarskite-yttro-

columbite series (or the yttrocolumbite-yttrotantalite series) have rare

earth assemblages characterized by dominant Y and high Dy andf ot Er

andf or Yb (Butler, 1958); generally, these minerals have U>Th ii U

exceeds a few wt. 7o. In the series aeschynite-priorite, however, there is

some evidence that Th is high in aeschynite (where Ce earths predom-

inate) and U is high in priorite (where Y-earths predominate). The re-

examination of lyndochite and the remarks on the distribution of the

rare earths in davidite are relevant here (Butler, 1957b,1961; Butler and

Hall, 1960). Further, Semenov and Barinskii (1958) have shown that

thorite and thorianite are rich in the light lanthanons whereas uraninite

is rich in the heavy lanthanons (and presumably Y also)' In monazites,

then, the presence of Th in varying amounts does not act as a control or a

restraint on rare earth assemblages; the l ight lanthanons can all produce

the monazite structure perfectly well and the reason for lack of correla-

tion between Th and rare earth distribution is apparent. The presence in

monazite of the smaller sized Ua+ however, does tend to facilitate uptake

of the heavier lanthanons. Ua+, has a radius of 0.97 A compared with

0.97 A for Gd, 0.92 A for Dy3+ and.1.02 A for Tha+.In the Iight of these observations, it wil l be of interest to see if future

work shows that Th conten! in xenotime has a greater effect in changing

typical rare earth distribution patterns than has U.

MALAYAN ALLUVIAL MONAZITE I22I

Pnysrcer, PnopBntrBs

Specific graaity: According to Palache et al,. (1951, p. 693) and Frondel(1958, p. 157) the specific gravity of monazite ranges from 4.6 to 5.4,lying mostly within the values 5.0 and 5.2. They further state that thespecific gravity "increases with content of Th." The atomic weights ofTh and U (232 & 238) appreciably exceed those of the lanthanons (162for Dy to 139 for La) and Y (89) and some correlation between ThOz(plus U3Os) content and specific gravity of monazite is to be expected.The Malayan samples, which vary from 5.06 (No. 7) to 5.23 (No. 15),show that the statement above, that the specific gravity increases withTh content, is an over-simplification. Sample No. la, with a total ofl l .5270 U and Th oxides, has a lower specific gravity (5.19) than sampleNo. 15, in which the specific gravity is 5.23 and the U and Th oxidestotal less than 7/6. The specific gravity of the Malay an rnonazites is bestcorrelated with the ratio of "l ight" to "heavy" oxides i.e. YlLa*Ce*Pr:Nd*Sm+Gd+Th*U (wt. 7o oxides). When plotted graphically,however, sample Nos. 1, 4, and 10 are rather off the main trend of in-creasing specific gravity with decreasing values of the oxides ratio, andthere is no obvious explanation. Karkhanavala and Shankar (1954)found, on heating, an increase in specific gravity oI a monazite due to themonazite being metamict. The specific gravities of the Malayan mon-azites were re-determined after heating at 1150o C. for f ive hours but,although they all increased between 0.04 and 0.10 above the originalvalues, the general trend and anomalies remained the same. In some ofthe samples heating produced a change in color, the most striking beingNo. 1a which changed from bright apple-green to a dull green with a dullrust-brown coating, and No. 1 which changed from clear yellow to abright clear red.

Mass magneti.c susceptibility: Richartz (1961) has recently studied mag-netic fractions of a monazite from Brazil in which the susceptibilityranged from near that of ilmenite to near that of (associated) zircon. Thevariation in magnetic properties was well correlated with a change incomposition of rare earths; Fe in the monazite was below 0.1/6 andtheelements Mg, Ca, Ti, Mn, Cu, and Eu were present as traces only. Theatomic percentages of the individual rare earths in total rare earthschanged for the six successively less magnetic fractions as follows: Lafrom 19.7 to 28.8/6, Ce from 41.3 to 50.3Ta, Nd from 2I.4 to l l.9/s, Smfrom 5.6 to t.5/6, Gd from 3.9 to 2.8/6, and Y from 3.5 to below l/6.The oxides of these elements show contrasting susceptibilities; LazOa*YrOa are diamagnetic, Ce2O3, NdzOa, and SmzOa are weakly to moder-ately paramagnetic, Gd2O3 is strongly paramagnetic, and DyzOa verystrongly paramagnetic.

1222 B. H. FLINTER, J. R. BUTLER AND G. M. HARRAL

The variation in mass magnetic susceptibility of the Malayan mon-

azites is large. The relationship between magnetic susceptibility and rare

earth composition of these monazites, although not clear-cut, follows

broadly Richartz's f indings. Monazite No. 1a, with its unusually high

Dy content, is clearly the most magnetic; the high U content (516,being

the only sample with more than 1y'6 U3Os) may also be a contributing

factor, but it is not l ikely to have as great an effect as the Dy. Slightly

Iess magnetic is monazite No. 15 which has the highest Gd content. Both

these monazites are considerably poorer in diamagnetic La than the

others. At the opposite end of the scale are samples 14, 7 , and 8. These

are, together with Nos. 1 and 10, highest in La' Nos. 8 and 13 are also

lowest in Gd.The Malayan samples indicate that similarity in mass magnetic sus-

ceptibil i ty does not necessarily mean similarity in composition. Mon-

azites 7 and 8 are in fact similar in both, but 9 and 10, though similar in

susceptibil i ty, are not similar in composition. Nor does the reverse hold:

samples 5 and 12 are similar in composition' but differ considerably in

susceptibility.The presence of unsought-for Fe in the samples (except for No. 1a,

which had less than tTd,^uy be the reason for the lack oI a clear-cut

relationship, but this is not considered to be the case. It appears far

more likely to be owing to the compensating compositional effects of the

highly contrasting rare earth elements.

X-ray powd,er pattern: Powder photographs of eight of the sixteen sam-

ples were taken, using a camera of 114.83 mm. diameter with Ni-fi l tered

copper radiat ion, CuKa: 1.50450 A.The ionic radii, after Ahrens (1952), of the rare earth elements are

centered around 1.05+0.03 A. Wittt itt this range lie Ce (1.07), Pr (1'06),

and Nd (1.04), and also Th (1.02). Sm, Gd, and U are slightly smaller at

1.00, 0.97, and 0.97 A respectively while Y is considerably smaller and

La considerably larger at0.92 and 1.14 A respectively. This wide range

in ionic radius could reasonably be expected to reflect a corresponding

variation in the interplanar (d-) spacings of the mineral, and it is logical

to assume an increase in the interplanar spacing with a predominance of

La and a decrease if the smaller ions (Sm, Gd, U' and Y) predominate'

Molloy (1959) states (p. 517) that variation in rare earth composition

does cause a small variation in cell dimensions, but he does not give

details. In his conclusions (p. 531) he goes on to say that "the range in in-

dividual (hkl) reflections . . . does not yield any simple relation to

chemical composition." Richartz (1961) found that the lines of an un-

differentiated monazite concentrate were diffuse whereas those of the

MALAYAN ALLUVIAL MONAZITE 1223

fractions separated from it were quite sharp. This he concluded to be dueto the greater compositional variation (in the La* Ce/Nd* Sm* Gd* Yratio) in the undifferentiated concentrate giving rise to a spread of latticevalues.

The patterns obtained on the Malayan samples conformed to the gen-eral pattern given in the ASTM powder data card 4-0612 and by Frondel(1958, p. 158, after Karkhanavala and Shankar) but the actual valuesdifier. The data for the Malayan samples do, in fact, indicate that for d-values below, and including, the (200) reflection (above which value the

Telr,n 4. Svsrnnr.qrrc VenrerroNs ol X-Rav DrllnncnoN SpacrNcs (A)

WrrnrN tso RaNcn hkl :zN ro 040 Cu RrornrroN, Ni Frr- : ren X:1.50450 A

15ta

t312I

7487

ASTM 4-0612Frondel (1958)

212

1 . 9 41 .951 .951 .951 .951 . 9 51 . 9 6t . 9 61 . 9 7t . 9 7

210120 031

3 . 2 43 . 2 63 . 2 53 . 2 53 . 2 53 . 2 53 . 2 73 . 2 73 .313 .29

3 .053 .05

f

2 . 9 42 . 9 52 . 9 42 . 9 52 . 9 42 . 9 4+

2 . 9 92 . 9 8

2 . 8 32 . 8 22 . 8 42 . 8 42 . 8 32 . 8 32 . 8 32 . 8 42 . 8 82 . 8 7

2 . 1 62 . 1 62 . 1 72 . 1 82 . 1 72 . t 72 . r 82 . 1 82 . r 92 . t 9

1 . 8 51 . 8 51 .861 .861 .861 .861 .861 .86I . 8 8r . 8 7

-L

t . , J

1 . 7 3I ' J

l . / J

t . 7 3-r

r . 7 41 . 7 51 . r J

33 .063 .063 .063 .063 .063 .093 . 1 0

f Line present but not measured.

accuracy ofmeasurement from photographs is unreliable), a definite trendof variation occurs (see Table 4), particularly for reflections whered:3.24-3.27, d:2.16-2.18, and d:1.9a1.96, and there is a reasonablysympathetic relationship between this variation and the La:(Smf Gd+U+Y) ratio. Monazite Nos. 15 and 1a, with the lowest d-values(smallest spacings) and the lowest La content, seem to indicate that Lahas a slightly greater effect than the group of smaller ions. They also con-firm the ineffectiveness of Nd, as they contain the highest and lowestNd content of all the samples.

CoNcrustoNs

Although certain trends are indicated for the properties studied, nosimple correlation with composition can be established. The position isunfortunately complicated by the effects of compensating replacements,such as the presence of equal amounts of a large or heavy element andcorrespondingly small or light element, which counter the indicated

B, II. FLINTER, J. R. BUTLER AND G. M. HARRAL

Teslt 5. Lrsr ol Loclr,rrrcs or rnn Sllpr.ns Srrmruo

No. State Locaiitl'

Siau Hin Mine, SemelingSiau Hin Mine, SemelingLarut Tin Fields Ltd., TaipingWah Sung Tin Mine Ltd., TrongS.K.C. Ltd, (Kinta), Batu GajahTronoh Mines Ltd., KamparHung Soon Kongsi, BidorAustral Amalgamated Tin Ltd., SungkaiKarak Tin Mine, KarakSg. Way Dredging Ltd., Sg. WayPetaling Tin Ltd., PetalingTambah Tin Dredging Co. Ltd., PertangTongkah Compound No. 2, N.L., SerembanRantau Tin Dredging Co. Ltd., SiliauPulau BesarBakri Mining Co-, Bakri

G.S.F.M.No.

Il a

4

67

8I

101 1t2I J

1 A

l . )

KedahKedahPerakPerakPerakPerakPerakPerakPahangSelangorSelangorN. SembilanN. SembilanN. SembilanMalacca

Johore

E-2/rE-2/laE-8/2E-8/16E-8/4F,-8/7E.8/eE-8/14E.6/rF.-rc/2F.-r0/rE-s/2E-s/3E.-s/rE-4/rE-r/r

trends, giving anomalous results. The indicated effects imposed by theindividual elements are as fotrlows:

Ce, La, Nd, Gd, Sm, U, and Th collectively affect the specific gravity; a high value ofCe{La gives a low specific gravity whilst a high value of Nd*Gd*Smf UfTh givesa high specific gravity.

La, Gd, and Dy afiect the mass magnetic susceptibility; U appears to do so. A composi-tion of high Gdf Dy (and possibly U) and low La content gives a high magnetic sus-ceptibility and vice versa.

La has a definite (the most important) efiect on the interplanar and unit cell dimensions,increasing them both. A decrease in both is caused by Y, U, Gd and, to a lesser extent, Sm.

Because of its very limited variation, little can be concluded about Pr. It probably hassome effect on the specific gravity; and has apparently no efiect on the magnetic sus-ceptibility or structural dimensions.

Th, though the most variable has virtually the least effect. It has no efiect on the magneticsusceptibility or on the structural dimensions and only afiects (increases) the specificgravity in combination with U, Sm, Gd, and Nd.

In addition to its effect on the physical properties, indicated above, a high U contentaffects the lanthanon distribution in favor of the heavy lanthanons.

Acr<NowrnoGMENTS

Grateful acknowledgment is made by B. H. Flinter and G. M. Harralto J. B. Alexander, Director of Geological Survey, Federation of Malayafor his encouragement and for permission to publish this paper and to thestaff of the Geochemical Division for help in the various chemical deter-minations. Thanks are due, on behalf of J. R. Butler, to the Director ofRothamsted Experimental Station for the use of the Mannkopfi glass

MALAYAN ALLUVIAL MONAZITE

prism spectrograph for determining the lanthanons and to Alan Z. Smith

(Imperial College) for determining Y by r-ray fluorescence.

RnrrnrNcns

AnnnNs, L. H. (1952) The use of ionization potentials. Part L Ionic radii of the elements.Geochi.m. C osmochi.m. Acta, 2, 155-169.

Bonovsrcv, I. B. ,lro V. I. Gnnasruovxsv (1945) Rare earths in minerals. C.R. (Dokl'ady)Acad.. Sci. U.S.S.R. 49, 353-6.

Butr,nn, J. R. (1957a) The spectrographic determination of the rare earths. Spectrochim.Acta,9,332-340.

- (1957b) Rare earths and thorium in lyndochite. Am. Minera.l., 42, 671-5.-- (1958) Rare earths in some niobate-tantalates. Mineral. Mag. 31,763-780.--- (1961) The constitution of davidite. Econ. Geotr. 56,4424.--- AND P. G. EMsREv (1959) Delorenzite is tanteuxenite. Minerol,. Mag.32,308-313.

Rosnulnv Har.r. (1960) Chemical characteristics of davidite. Econ. Geol'.55,1541-1550.

CannoN, M. K., C. R. Nlnsnn, H. J. Rosr, JR. AND F. A. Hrr.rrnnmqn (1958) Fractionalprecipitation of rare earths with phosphoric acid. [/. S. Geol. Swttey Bu'11. 1036-Nt253-275.

l'noNoEt-, Cru'rono (1958) Systematic mineralogy of uranium and thorium. U. S.Geol'.Suntey Bull,. 1O64.

Gunsr, R. J. mo J. B. ZruuntueN (1955) Determination of uranium in uranium con-centrates, use of ethyl acetate. Anal. Chem.27r93l-6.

Hrrorl, R. H. ,lNo V. A. F.lssrr, (1958) X-ray fluorescent spectrometric determinationof yttrium in rare earth mixtures. AnaL Chem. SO, 176-t79.

HnrNnrcrr, E. Wu., R. A. Bonup amn A. A. LovrNsoN (1960) Relationships betweengeology and composition of some pegmatitic monazites. Geochi'm. Cosmochim. Acta,19,222-237.

ILmtrn,lno, F. A., M. K. C.sRnoN mo H. J. Rosr, Jn. (1957) Re-examination of rhabdo-phane (scovillite) from Salisbury, Connecticut. Bull. Geol. Soc. Am.,68' 1744.

Isuarr., A. M. mro H. F. Henwooo (1937) The use of hexamine for the separation ofthorium lrom the rare earths, and its application to the determination of thorium inmonazite sand.. Ana.l,yst, 62, 185-191.

Kemnnrever,A, M. D. .tNo J. Sruxran (195a) An x-ray study of natural monazite.India Acad.. Sci. Proc.40A,67.

Mor.r.ov, M. W. (1959) Comparative study of ten monazites. Am. Mineral.44, 510-532.Munare, K. J., C. V. Durne, M. Tnrxrrne DA CosrA, AND J.J.R. Bn.c,Nco (1958) Com-

position of monazites from pegmatites in eastern Minas Gerais, Brazil. Geochim. Cos-mochim. Acto, 16, 7-74.

Munarl, K. J., H. J. Rosn, Jn., aNo M. K. C.q,nnoN (1953) Systematic variation of rareearths in monazite. Geochi.m. Cosmochim. Acta,4, 292-300.

J. J. Gress (1957) Systematic variation of rare earth elements in cerium-earthminerals. G eoch.i.m. C o s m o c him. A c ta, ll, 14l-161.

Peucnn, C., H. Brnu,lN alro C. FnoNnu. (1951) Dono's System oJ Mi.neral'ogy. 7th ed.,Vol. II. John Wiley & Sons, New York.

Rrcnanrz, W. (1961) Uber kristallchemische Untersuchungen und magnefische Auf-bereitung von Monazit. Fortschr. Minerol,.39, 53-59.

Rosr, H. J., Jn., L. V. Br,eor nxr M. Ross (1958) Earthy monazite at Magnet Cove,Arkansas. Am. Mined,. 43,995-7.

T226 B. H. FLINTER, T. R. BUTLER AND G, M. HAKRAL

SrunNov, E. I. aNo R. L. B.q,nnsrrr (1958) Compositional characteristics of rare earths inminerals. Geokhimiyo, 1958, 314-333 (translated, Geochemi.stry, 1958, 398-419).

VAnsrtnrx, E. E., A. I. Tuce,nnov auo N. V. TtrneNsreva (1955) The distribution ofrare earths in monazites. Doklody Acad. Nauk. S.S.S.R. lO4,268-271.

- (1956a) Distribution of the rare earths in monazites of granitoids. Doklod,y Aca.d.trfloaft. S.S.S.R. 106, 691-2.

-- (1956b) On regularities in the distribution of rare earths in some minerals. Gea-hhi.rniya, 1956, 3G56.

Wvr-r, A. W. (1950) Composition of some Australian monazites. Austral. four. Appl. Sci.t, 1964-171.

Manuscri.pt recei.ued, February 26, 1962; aecepted. for publ,i.cation, September 5, 1963.