Clay Minerals (1981) 16, 383-393.

A ME T H O D TO E L I MI N A T E T HE B A C K G R O U N D I N
X - R A Y D I F F R A C T I O N P A T T E R N S OF O R I E N T E D
C L A Y MI N E R A L S A MP L E S
S. J. VAN DE R GAAS T AND A. J. VAARS
Netherlands Institute f or Sea Research, P.O. Box 59, Texel, The Netherlands
(Received 15 June 1981)
ABSTRACT: A method is described for calculating, and then subtracting, the background from
X-ray diffraction patterns of oriented clay mineral samples. Ti-K~ radiation is used and, to
minimize the absorption of this radiation by air, a vacuum and helium-flushed device has been
developed. This device can be used with other X-ray sources, offering a considerable increase of
intensity---e.g. Co-Kc~ radiation is increased by 125~. With the background-eliminated patterns a
better semi-quantitative estimate of the composition of clay mineral mixtures is possible. Small
differences in composition of two samples can be identified by subtracting one of the
background-eliminated patterns from the other. Using this method, peak maxima of smectite-
group minerals can also be accurately determined.
I n or der t o est i mat e t he cl ay mi ner al compos i t i on of sedi ment or soil sampl es, the
intensities of some l ow-angl e basal reflections of cl ay mi neral s on t hei r X- r ay di ffract i on
pat t er ns are usual l y measur ed. At these l ow angles, the reflections are super i mposed on a
shar pl y decreasi ng backgr ound t hat has to be subt r act ed f r om the reflections. To i mpr ove
this met hod, det er mi nat i on of t he backgr ound is necessary.
I n nor mal X- r ay pat t er ns measur ed wi t h a sui t abl e di vergence slit (which pr ovi des a
const ant r adi at ed vol ume at all angles), t he backgr ound is compos ed of (Fig. 1) - - ( a)
i ncoherent radi at i on: Co mp t o n di ffusi on (3), fl uorescent r adi at i on (4) and t emper at ur e-
diffuse scat t er i ng (3) (Cullity, 1967); (b) appar at us - dependent fact ors: i nci dent beam (5)
and ai r scat t er (6); (c) t ot al reflection (7) (Gui ni er, 1956). The reflections in the X- r ay
pat t er n caused by coher ent r adi at i on resul t not onl y f r om the 00l reflections of the cl ay
mi neral s but al so f r om di ffract i on by small part i cl es (1), mi cr o- por es (1) ( Tehoubar &
M6ring, 1969) and cryst al defects (2) (Pons, 1980; Pons et al . , 1980). Al t hough t he l ast
t hree effects cannot strictly be r egar ded as backgr ound, t hey oft en are r egar ded as such
because t hey cannot be di st i ngui shed f r om t he real backgr ound. The backgr ound effects
ment i oned above are al so influenced by t he Lor enz- pol ar i zat i on f act or (6), the abs or pt i on
f act or (4) and the at omi c scat t eri ng f act or (8). On X- r ay di ffract i on traces, t he sum of t he
effects 1 t o 8 results at l ow angles in a curve mor e or less represent ed by curve 9 of Fig. 1.
On X- r ay pat t er ns obt ai ned wi t h Co- Ks r adi at i on, the t rue backgr ound is oft en onl y
visible f r om ,-~ 16 ~ 20. To est i mat e t he backgr ound on the l ow-angl e side of t he 00l
reflections of cl ay mi neral s in X- r ay di ffract i on pat t er ns we chose Ti - Ka r adi at i on
(2 = 2.74973 A). Meas ur ed at 50~ relative humi di t y (r.h.), t he 001 smect i t e reflection t hen
gave its ma xi mum at ~ 11 ~ 20. Thi s peak shift makes par t of the backgr ound on t he
l ow-angl e side of t he pat t er n visible. To obt ai n a sui t abl e di ffract ed i nt ensi t y at t he angl es
0009-8558/81/1200-0383502.00 9 1981 Blackwell Scientific Publications
384 S. J. van der Gaast and A. J. Vaars
Ze 4 0
FIG. 1. Factors contributing to 'background' in normal X-ray diffraction patterns. It is assumed
that the irradiated volume is kept constant at all angles and the total reflection peak is overlapped
by the incident beam. (1) small particles, micro-pores; (2) crystal defects of large spacing minerals;
(3) Compton diffusion, temperature diffuse scattering; (4) fluorescent radiation, absorption
factor; (5) incident beam; (6) Lorenz-polarization factor, air scatter; (7) total reflection; (8) atomic
scattering factor; (9) 'background' resulting from the combined effects 1-8.
at which the clay mineral reflections are measured, and also to avoid the influence of the
incident beam on the pattern, an automatic divergence slit system (ADSS) was used. This
device varies the divergence continuously, so that, whatever the angle of the goniometer,
the same area is irradiated; this results in a decreasing irradiated volume with decreasing
angle. Thus with decreasing angle there is: (a) a progressive decrease of the temperature-
diffuse scattering; (b) a slowly increasing Lorenz-polarization factor effect; (c) a decrease
in the absorption factor effect; (d) a slight decrease in the atomic scattering factor effect.
This ' ADSS effect' is also accompanied by a decrease in the diffracted intensity from the
small particles, micro-pores and crystal defects. With the met hod described in this paper,
Compt on diffusion and fluorescent radiation were prevented from reaching the counter
by using a crystal monochr omat or between the sample and the detector. Air-scatter was
avoided by flushing the specimen chamber with helium.
To determine the form of the background curve, we analysed the X-ray patterns of a
kaolinite, two smectites of different crystallinity, some size fractions from two sediment
~ON0~OMATOR CR~ST~L
I 0ETECTOR
?/2
/
J
~CUUM
FiG. 2. Schematic view of the vacuum-helium device.
Background elimination in XRD patterns 385
s a mp l e s a n d a c e r a mi c t i l e s a mp l e h o l d e r . Us i n g a c o mp u t e r , t he b a c k g r o u n d wa s
c a l c u l a t e d f or al l p a t t e r n s f r o m t he f o r mu l a y =Ax B. The b a c k g r o u n d va l ue s wer e
s u b t r a c t e d f r o m t he or i gi na l p a t t e r n a n d t he b a c k g r o u n d - e l i mi n a t e d p a t t e r n p l o t t e d .
I f t he p e a k t o b a c k g r o u n d r a t i o i s l ow i n n o r ma l X- r a y p a t t e r n s , t he p e a k va l ue s a r e
of t e n s hi f t ed t o hi ghe r A. val ues . Wi t h t hi s me t h o d mo r e a c c u r a t e p e a k va l ue s wer e
o b t a i n e d .
Becaus e o f t he hi gh l evel o f a b s o r p t i o n o f Ti - K~ r a d i a t i o n b y ai r , a devi ce was
d e v e l o p e d i n whi c h a p a r t o f t he X- r a y p a t h i s e v a c u a t e d a n d a n o t h e r p a r t is
he l i um- f l us he d ( Fi g. 2). Wi t h t hi s devi ce t he i nt e ns i t y o f t he Ti - K~ r a d i a t i o n wa s i nc r e a s e d
200 t i mes . An i nc r e a s e o f t hi s ma g n i t u d e wa s ne c e s s a r y t o o b t a i n a wo r k a b l e p a t t e r n .
MAT E R I AL S AND ME T HODS
Th e b a c k g r o u n d was d e t e r mi n e d on < 2 / ~ m f r a c t i o n s o f a k a o l i n i t e f r o m Pr ove nc e ,
Fr a n c e , a mo n t mo r i l l o n i t e f r o m Wy o mi n g , US A, a s me c t i t e f r o m Et h i o p i a , 2- 0. 5,
0. 5~) - 2, < 0-2 a n d < 0. 05 # m f r a c t i ons f r o m a s e d i me n t 50 c m f r o m t he t o p o f a c or e f r o m
t he No r t h At l a n t i c , < 2 # m f r a c t i ons f r o m t hi s s e d i me n t a n d al s o f r o m a s e d i me n t 25 c m
f r o m t he t o p o f t he s a me cor e, a nd, f i nal l y, a c e r a mi c t i l e s a mp l e hol de r . The s e d i me n t a n d
s me c t i t e f r a c t i ons wer e Ca - e x c h a n g e d . Or i e n t e d s a mp l e s wer e p r e p a r e d on p o l i s h e d
p o r o u s c e r a mi c t i l es ( Di i mml e r & Sc hr 6de r , 1965) a n d d r i e d a t r o o m t e mp e r a t u r e . X- r a y
r uns wer e c a r r i e d o u t a t 5 0 ~ r . h. a n d al s o a f t e r gl yc ol a t i on. Th e r . h. wa s k e p t c o n s t a n t
wi t h a h u mi d i t y g e n e r a t o r d e v e l o p e d a t o u r i ns t i t ut e. An X- r a y g e n e r a t o r P W 1730 wi t h
TABLE t. Apparatus and experimental conditions.
Broad focus Co-tube, Broad focus Ti-tube,
X-ray tube PW2256/ 20 (Philips) PW2250/ 22 (Philips)
Rating
Window tube
Divergence slit
Receiving slit
Anti-scatter slit
Monochromator
Counter
Counter window
X-ray path
Window vac. part
Specimen chamber
flushed with
Scanning speed
Scanning range
Scanning time
Recording
40-50 kV and 40-50 mA 25 kV and 48 mA
Be, 300 pm Be, 150 #m
0.5 ~ ADSS PW1386/50 (Philips)
open at 0.3 ~ 20
0-2 mm 0.2 mm
1 o 1 o
Graphite, AMR Graphite, AMR
Proportional, Proportional krypton
PW1965/60 (Philips) counter with a signal
amplifier, 60 x
Mica Be, 150 pm
Vacuum (1 mm Hg) Vacuum (1 mm Hg)
6 #m Mylar foil 6/~m Mylar foil
Air, 50~ r.h., 150 ml Helium, 50~ r.h., 150 ml
min - l rain - 1
0.5 ~ 20 mi n- 1 Stepscan, 4 sec 0.02 ~ 20 l
3-16 ~ 20 0-30 ~ 20
26 min 140 min
Recorder PM 8 2 0 3 Computer, Tektronix 4051,
(Philips) floppy disc unit (File
Manager Tektronix) and plotter
386 S. J. van der Gaast and A. J. Vaars
channel control PW 1390 and motor control PW 1394 (Philips) was used; apparatus
constants are given in Table 1. The Simple Linear Regression Program ( St at i st i cs 1,
Tektronix) was used to estimate the shape of the background.
RESULTS
Using the reflection method in combination with the ADSS implies that the irradiated
sample volume decreases with decreasing angle. Because small particles, micropores and
crystal defects show their diffraction effects at low angles, they will only slightly affect the
part of the X-ray diffraction pattern studied. To eliminate any interference from these
low-angle diffraction effects, a well-ordered kaolinite sample was chosen as a starting
point for determining the background. Applying the ADSS, the graphite monochromator
and the vacuum-helium device, the background on this sample comprises only the
temperature-diffuse scattering, the absorption factor, the Lorenz-polarization factor and
the total reflection, influenced by the ADSS effect.
G F
L . 9
FIG. 3. Background curves calculated from X-ray diffraction pattern values, represented by dots,
compared with curves calculated from the formula y = A x s. A = kaolinite, Provence; B = ceramic
tile; C=mont mori l l oni t e, Wyoming; D=smect i t e, Ethiopia; E=2- 0. 5 #m, F=0. 5- 0. 2 pm,
G= <0.2/~m, H = <0.05/~m sediment fractions; I = <2 pm sediment fraction (E-I all from 25
cm depth in Nort h Atlantic core); J = < 2 #m sediment fraction from 50 cm depth in the same core.
The values shown are R 2 values, which reflect the fit of the calculated curve with the measured
values.
Background elimination in XRD patterns 387
For smect i t e- cont ai ni ng sampl es, the r.h. st rongl y influences t he effects caused by 1 and
2 in Fig. 1. Bot h l ow and hi gh humi di t i es resul t in an i ncrease of the di ffract ed i nt ensi t y in
the 1-4 ~ 20 region.
Thi s was not f ound wi t h t he kaol i ni t e sampl e and onl y a smal l devi at i on was obser ved
f r om 1-2. 5 ~ 20 when it was gl ycol at ed. Gl ycol at i on of t he smect i t es and the smect i t e-
cont ai ni ng sedi ment sampl es mas ked t he backgr ound bet ween the 001 peak and 1 ~ 20.
Because this par t of the pat t er n is very i mpor t ant f or obt ai ni ng backgr ound values, we
deci ded t o car r y out t he meas ur ement s f or est i mat i ng the backgr ound at 50Y/o r.h. Under
these condi t i ons the backgr ound is least obscur ed by the coher ent r adi at i on caused by the
cl ay mi neral s. The backgr ound meas ur ement s were st art ed at 0 ~ 20, al t hough bel ow 1 ~ 20
the measur ed val ues are influenced by i naccur at e openi ng of the ADSS. I n the pat t er ns of
the > 0.05 #m sedi ment f r act i ons in Fi gs 5, 6 and 7 and t hat of the smect i t e in Fig. 5, the
t ot al reflection peak can be seen at a bout 1 ~ 20 where the backgr ound is not el i mi nat ed. I n
some cases a peak, caused by a par t of t he i nci dent beam, can be seen at about 0.5 ~ 20 (in
Fig. 5, 2-0. 5 #m sedi ment fract i ons; in Fig. 6, < 0.2 and < 0.05/~m sedi ment f r act i ons and
the Et hi opi an smectite). The peak caused by the t ot al reflection is par t l y i ncl uded in the
backgr ound measur ement s. The i nci dent beam had no influence on it.
The f or mul a y = Ax 8, in whi ch x and y r epr esent the coor di nat es of the measur ed val ues
in the di agr ams and A and B are const ant s, appear ed t o descri be the backgr ound mos t
accur at el y (Fig. 3).
sm
~4s4~
sm
l/Mic~ j
995
' 2 e . . . . . 7 co K a ' 2 e . . . . . i o r co Km
FIG. 4. X-ray diffraction patterns of the sediment fractions and the Ethiopian smectite, run with
Co-K~ radiation at 50~ r.h. (left) and after glycolation (right). Chl = chlorite, K =kaolinite,
I = illite, Sm = smectite.
388 S. J . van der Gaast and A. J . Vaars
Fi g. 3, c ur ve A, s h o ws t he ve r y g o o d fit o f t he di f f r ac t i on c ur ve o f t he ka o l i ni t e s a mpl e
wi t h t he c ur ve c a l c ul a t e d f r o m t he me a s u r e d val ue s . A near pe r f e c t fit wa s a l s o f o u n d f or
t he e mp t y c e r a mi c ti l e ( Fi g. 3, c ur v e B). In t hi s c as e we we r e abl e t o me a s u r e t he
b a c k g r o u n d f r o m 0-7 ~ 20, pr o ba bl y due t o t he ve r y fl at s ur f ac e o f t he ti l e. At angl e s
> 6 ~ 20, t he b a c k g r o u n d i nc r e as e d be c a us e o f t he t e mpe r at ur e - di f f us e s cat t eri ng, t he
a b s o r p t i o n f a c t o r a nd t he a t o mi c s c at t e r i ng f a c t o r i nf l ue nc e d by t he A D S S ef f ect . I n t hi s
c ur ve we me a s u r e d t he b a c k g r o u n d b e t we e n 0. 7 ~ a nd 6 ~ 20. The s l i ght b a c k g r o u n d
i nc r e as e wa s a l s o o bs e r v e d i n t he X- r a y pat t e r ns o f Fi gs 5, 6 a nd 7 but c o u l d be i g no r e d i n
I
I l i i : I /
I { 7 " , , i "~.i/
/ " ~ J !
/ i j ' /
15741
b
I
Sm
14G6~
Chl 11714 ~ 9 S 5 ~
i l I~i ,
I
I' ,I
/ "
,! I
J 1 4 ~ 8 ~ I t
::_ ,j"W' -
<o05~
/ f . . . . . . " r <j /
. . . . . . . . . . . . . . . . W / ~\
26' 22 IO Ti Kc~ 0 2(~ if2 I0 Ti KO( 0
FIG. 5. X- ray di ffracti on pat t erns o f the sedi ment f ract i ons and the Et hi opi an smecti te, run wi t h
Ti-K~ radi at i on and the vac uum- he l i um devi ce at 5 0 ~ r. h. bef ore (left) and after (ri ght)
background el i mi nat i on. Chl = chl ori te, K = kaol i ni t e, I = illite, Sm = smecti te.
Background elimination in XRD patterns 389
b a c k g r o u n d c a l c ul a t i on. Wi t h de c r e a s i ng pa r t i c l e si ze a s ma l l e r p a r t o f t he X- r a y p a t t e r n
c o u l d be us ed f or b a c k g r o u n d val ues , due t o t he i nc r e a s e i n wi dt h o f t he s me c t i t e p e a k a n d
a n i nc r e a s i ng a mo u n t o f t hi s mi n e r a l i n t he s a mpl e . Th u s one b a c k g r o u n d va l ue be t we e n
3 ~ a n d 4 ~ 20 on each o f t he cur ves E, I a n d J ( Fi g. 3) s h o we d a s ma l l d e v i a t i o n b e c a u s e o f
t hi s. The c a l c u l a t i o n o f c ur ve H ( Fi g. 3) f r o m onl y t he f o u r va l ue s 1-8, 2. 0, 2-5 a n d 26 ~ 20,
yi e l de d a ve r y g o o d R 2 va l ue ( t he R 2 va l ue s r ef l ect t he fi t o f t he c a l c u l a t e d c ur ve t o t he
me a s u r e d val ues ) . Ho we v e r t he X- r a y p a t t e r n f r o m t hi s s a mp l e s hows a p e a k a t a b o u t 1.4 ~
lIMl~
9 9 5 1
ehl
sm
7 84
2 e 2~ I o Ti K ~ 0 2 ~ 22 I 0 T i K ~ 0
FIG. 6. Same samples as in Fig. 5 after glycolation before (left) and after (right) elimination of the
background. Chl = chlorite, K = kaolinite, I = illite, Sm = smectite.
390 S. J. van der Gaast and A. J. Vaars
I / M i c :
9 9 5 ~
c.p
~o41 7~K41
I
I "
I
T 4 ~ e A ,~
/,I
/
2e 22 io Ti K~ 0 2e 22 I0 ' ' TI KO~ 0
FIG. 7. X-ray diffraction patterns of the < 2 #m sediment fractions from 25 and 50 cm depth run
with Ti-Ka radiation and the vacuum-helium device at 50~ r.h. Left side before and right side
after background elimination. The vertical scale of the background-cleared patterns was
'equalized' with the 001 kaolinite reflections and subtracted.
20, which slightly obscures the 1 ~ 20 backgr ound value. This peak, caused by coher ent
r adi at i on f r om small particles, mi cr o- por es and cryst al defects, cannot be regarded as
bel ongi ng t o the backgr ound. In such a case the 1 ~ 20 val ue has to be used as this
appr oaches the real backgr ound much closer t han the 1.8, 2.0 and 2.5 ~ 20 values.
Figs 5 and 6 show the pat t erns of the sediment samples and the Et hi opi an smectite
obt ai ned with Ti-K~ r adi at i on and the vacuum-hel i um device, with and wi t hout
backgr ound. The Ti-K~ radi at i on, gives much bet t er peak resol ut i on, as can be seen by
compar i ng t he pat t erns in Fig. 4.
The result of subt ract i ng the backgr ound- el i mi nat ed pat t erns of t wo < 2 #m sediment
fract i ons is shown in Fig. 7. Peak val ue correct i ons are shown in Tabl e 2. Usi ng Ti-K~
r adi at i on wi t h the whol e vacuum-hel i um device, the i nt ensi t y was increased 200 times.
Co- Ks radi at i on, in combi nat i on wi t h the vacuum par t of the device, showed an intensity
increase of 90~. When t he ot her par t was si mul t aneousl y flushed with helium, the
i nt ensi t y increase was 125~o. The i nt ensi t y of Co-K/~ radi at i on increased about 100~
using bot h vacuum and helium. The specimen chamber can also be evacuat ed or flushed
with ot her gases, pr ovi ded t hat t hey do not at t ack the Myl ar windows.
The scanning time for one sample is 140 mi n (Tabl e 1). This l ong scanni ng t i me is due
to: (1) t he stepscan met hod, because f or comput er calculations t he count s have to be in
Background elimination in XRD patterns
TABLE 2. Angstrom values of the smectite peaks from the sediment
samples and the Ethiopian smectite run Co-Ks radiation and Ti-Ke
radiation, zx = values wi t h background, - = values without
background.
Co-Ks radiation Ti-K~ radiation
50~ r.h. glycolated 50~ r . h . glycolated
A / k / x - - A - -
2-0.5pm 14.66 17.69 14. 66 14.66 17.72 17.72
0.5-0.2/~m 14.64 17.69 14. 64 14.64 17-66 17.66
<0.2/~m 14.98 17.69 14. 57 14.54 17.70 17.66
<0.05 #m 15.25 18.39 14. 69 14.58 18.11 18.02
Smectite, Ethiopia 16.50 21. 38 15. 74 15.68 19.83 19.54
<2 pm, 25 cm 14.66 17.69 14. 66 14.66 - - - -
<2/~m, 50 cm 14.70 17.72 14. 66 14.66 - - - -
391
di gi t i zed f or m; (2) t he n u mb e r o f c ount s r equi r ed t o obt a i n a wor ka bl e di fference pat t er n;
(3) t he l ar ge s canni ng r ange ( f r om 0- 30 ~ 20). Because o f t he di gi t i zed s t or age o f t he c ount s
by t he c omput e r , t he X- r a y ge ne r a t or can c ons t a nt l y be used at a ma x i mu m r at i ng, i.e. t he
r at i ng is n o t a ny l onger c ons t r a i ne d by t he size o f t he X- r a y pa t t e r n i n r el at i on t o t he wi dt h
o f t he r ecor der paper .
DI S C US S I ON
Us i ng t he me t h o d descr i bed above, t he b a c k g r o u n d o f X- r a y pa t t e r ns f r o m or i ent ed cl ay
mi ner al sampl es can be cal cul at ed accur at el y f r o m t he f or mul a y = Ax ~, ut i l i zi ng t he
di f f r act i on cur ve val ues bet ween 1- 1. 5 ~ 20 and 26 ~ 20. The best resul t s ar e obt a i ne d when
t he sampl es ar e me a s ur e d at 50~o r. h. Wi t h sampl es c ont a i ni ng cons i der abl e a mo u n t s o f
ver y smal l part i cl es, t he 1 ~ 20 val ue o f t he di f f r act i on cur ve ma y be s o me wh a t hi gh,
r esul t i ng i n a smal l devi at i on o f t he cal cul at ed b a c k g r o u n d at t he l ower angl es.
Af t er t he b a c k g r o u n d is cal cul at ed it can be s ubt r a c t e d f r o m t he or i gi nal pat t er n. The
me t h o d was appl i ed t o t he s edi ment f r act i ons a nd t o t he Et hi opi a n smect i t e, r un at 5 0 ~
r. h. (Fi g. 5). As can be seen i n t he pa t t e r ns af t er b a c k g r o u n d el i mi nat i on, t he di f f r act ed
i nt ensi t y bet ween 1 and 5 ~ 20 i ncr eases wi t h decr easi ng f r act i on size. Thi s effect c a n be
caus ed by: (1) a decr ease i n par t i cl e size; (2) an i ncr ease i n t he a mo u n t o f mi c r o- por e s and
a decr ease i n mi c r o- por e size; (3) i ncr ease i n a mo u n t o f cr yst al defect s. The p h e n o me n a
me nt i one d u n d e r (1) a nd (2) shi ft t hei r refl ect i ons t owa r ds l ar ger angl es wi t h decr eas i ng
f r act i on size, r esul t i ng i n a di f f r act i on ma x i mu m at 1.8 ~ 20 i n t he < 0. 05 # m f r a c t i on o f t he
No r t h At l ant i c sedi ment . P h e n o me n o n (3) decr eases wi t h decr eas i ng angl e, due t o t he
ADSS effect. The cur ve o f t he Et hi opi a n smect i t e is near er t o t he basel i ne at t he s ame
l ow- angl e r egi on t ha n t he cur ve o f t he <0- 05 # m s edi ment f r act i on, a l t h o u g h t hei r
smect i t e pe a k size a nd pe a k wi dt h ar e a b o u t t he same.
Wi t h t he gl ycol at ed sampl es t hi s s ame i ncrease, s o me wh a t enhanced, c a n be obs er ved
(Fi g. 6). Du e t o gl ycol at i on, t he refl ect i ons caus ed by t he p h e n o me n a me nt i one d a bove
392 S. J. van der Gaast and A. J. Vaars
shifted. The smectite peak also shifted t owar d l ower angles, causi ng peak overl ap with the
low-angle maxi mum.
The advant age of eliminating t he backgr ound of X- r ay pat t er ns is shown in Fig. 7. To
calculate t he difference in clay mi neral composi t i on, the backgr ound- el i mi nat ed pat t er n
of the < 2/~m fract i on of the sedi ment f r om 25 cm in the core was subt ract ed f r om t hat at
50 cm and the difference curve, smoot hed over five measurement s, pl ot t ed (Fig. 7). Before
subt r act i on the vertical scale of the t wo curves was det ermi ned in or der t o equalize the
average val ue of the peak interval bet ween 22.30 and 22.42 ~ 20 (the 001 kaol i ni t e
reflection).
The resulting difference curve, assumi ng equal amount s of kaol i ni t e in bot h fractions,
shows t hat t he fract i on f r om 50 cm dept h cont ai ns mor e smectite t han t hat f r om 25 cm
depth. This curve shows also t hat a small peak remai ns at the low-angle side of t he
kaol i ni t e and illite/mica reflections. This is pr obabl y due t o a small difference bet ween t he
propert i es of these mi neral s in the t wo samples. The peak at 1.5 ~ 20 indicates a small
difference in the low-angle phenomena.
Due t o t he rapi dl y decreasing backgr ound, a smectite peak shift was observed with
decreasing particle size in the X- r ay pat t erns of the sediment samples obt ai ned with
Co-K~ r adi at i on (Tabl e 2). This shift was absent in the X- r ay pat t er ns obt ai ned wi t h
Ti -Ka radi at i on. Bot h with Co-K:r and Ti-K:r r adi at i on t he Et hi opi an smectite showed a
peak shift; this was compl et el y correct ed aft er backgr ound subt ract i on of the X- r ay
pat t er n obt ai ned with Ti-K~ radi at i on.
To check t hat the amount of kaol i ni t e was const ant in the t wo sediment samples, an
ori ent i ng i nt ernal st andar d was added. We obt ai ned encouragi ng results with mol yb-
denite, which with Ti -Ka r adi at i on gave its mai n peak at ~ 26 ~ 20. In this case the 27-28 ~
20 region was used f or backgr ound calculations. Mol ybdeni t e laas also been used as an
i nt ernal st andar d in quant i t at i ve X- r ay di ffract i on of clay minerals by Quaker naat (1970)
and Cody & Thomps on (1976).
CONCL US I ONS
A device has been devel oped to measure, at a few pr edet er mi ned poi nt s, the backgr ound
on the X- r ay pat t er n of an ori ent ed clay mi neral sample. Fr om t he measur ed backgr ound
values a backgr ound curve can be cal cul at ed with a f or mul a y = A x e and this subt ract ed
f r om the original X- r ay pat t ern. Slight differences in clay mi neral composi t i on can be
det ect ed by subt ract i ng one of the t wo backgr ound- cl ear ed pat t er ns f r om the ot her. The
advant ages of this appr oach are enhanced when an ori ent i ng i nt ernal st andard, such as
mol ybdeni t e, is used.
Due t o the rapi dl y decreasing backgr ound of X- r ay pat t er ns obt ai ned with Co-K~
r adi at i on the smectite peak somet i mes shifts to hi gher values. This effect is prevent ed by
using t he backgr ound el i mi nat i on met hod on the X- r ay pat t erns obt ai ned with Ti-K~
radi at i on.
The l ong wavel engt h Ti-K~ r adi at i on also gives bet t er resol ut i on. For all X- r ay sources
a consi derabl e gain in i nt ensi t y is obt ai ned when the vacuum-hel i um device is used. Kf l
lines f r om X- r ay sources can be used i f t he K~l~2 doubl et hi nders accurat e measurement s
of t he X- r ay reflections.
This met hod and device offer i nt erest i ng possibilities to clay mineralogists, sedimento-
logists and soil scientists interested in i mprovi ng t hei r quant i t at i ve analytical techniques.
Background elimination in XRD patterns 3 9 3
AC KNOWL E DGME NT S
Th a n k s ar e ext ended t o Mr F. Ei j ge nr a a m f or devi si ng t he c omput e r pr ogr a m; Mr F. Schi l l i ng f or as s i s t ance i n
c ons t r uc t i ng t he humi di t y gener at or ; Dr D. Spi t zer f or hi s hel p i n sol vi ng t he ma t he ma t i c a l pr obl ems ; Mr B.
Ve r s c huur f or dr a wi ng t he fi gures a nd Dr s D. Tc h o u b a r a nd C. H. Pons o f t he Uni ver s i t y o f Or l r ans , Fr ance, f or
t hei r s t i mul at i ng di scussi ons. The ma nus c r i pt was cri t i cal l y r ead by Dr s J. H. F. J a ns e n a nd D. Ei s ma a nd t yped
by Mr s J. Har t . The a ut hor s ar e especi al l y appr eci at i ve o f t he hel pf ul c o mme n t s o f t he s t af f o f t he X- r ay
appl i cat i on di vi si on, Phi l i ps, Al mel o a nd o f Dr W. H. Di emer o f I & S, Phi l i ps, Ei ndhoven.
R E F E R E NC E S
CODY R. D. & THOMPSON G. L. (1976) Clays Clay Miner. 24, 224.
CULLIa'Y B. D. (1967) Elements of X-ray Diffraction. Addi s on- Wes l ey, Readi ng, Ma s s a c hus e t t s . 514 pp.
Df3MMLER H. & SCHROEDER D. (1965) Z. Pflanzenernaehr. Dueng. Bodenk. 109, 35.
Gu Nmg A. (1956) Th~orie et Technique de la Radiocristallographie. Dunod, Pari s.
PONS C. H. (1980) P h D t hesi s, Or l r a ns Uni ver s i t y, Fr ance.
PONS C. H. , BEN BRAI-IIM J., Yf2CEL A. , TCHOUBAR D. & TCHOUBAR C. (1980) Clay Miner. 15, 111.
QUAKERNAAT J. (1970) J. sedim. Petrol. 40, 506.
TCHOUBAR D. & M~RING J. (1969) J. Appl. Cryst. 2, 128.
R I~ S U M I~: On dr cr i t une m&h o d e de cal cul d u f ond c ont i nu que l ' on pe ut ensui t e s ous t r ai r e du
clich6 de di f f r act i on X d ' u n 6chant i l l on d' ar gi l e or i ent r e. On ut i l i se le r a y o n n e me n t Ti - Ke ai nsi
q u ' u n e encei nt e 6t anche soi t s ous vi de soi t bal ayr e par l ' h~l i um afi n de di mi nue r l ' abs or pt i on de ce
r a y o n n e me n t par l ' ai r. Cet appar ei l l age pe ut ~tre ut i l i s6 avec d' a ut r e s s our ces de r a yons X
pe r me t t a nt un accr oi s s ement i mpor t a nt de l ' i nt ensi t 6 di f f r act r e recueillie, pa r exempl e l' efficacit~
du r a y o n n e me n t CoKe est accr ue de 125~. Les di a gr a mme s corri g~s de l eur f ond c ont i nu
pe r me t t e nt une mei l l eure es t i mat i on quant i t at i ve de la c ompos i t i on de m~l anges d' argi l es. Des
pet i t es di f f r r ences de c ompos i t i on de de ux 6chant i l l ons pe uve nt ~tre i dent i fi res en s ous t r a ya nt l ' un
des di a gr a mme s ai nsi cor r i gr s de l ' aut r e. En ut i l i sant celle m&h o d e les ma x i ma des rai es des
mi n r r a u x d ' u n gr oupe de smect i t es pe uve nt 6tre d&e r mi nr s avec pr r ci si on.
KUR Z R E F E R AT : Es wi rd ei ne Me t hode zue Er mi t t l ung u n d ans chl i egender Subt r akt i on des
Un t e r g r u n d e s von R6 n t g e n d i a g r a mme n or i ent i er t er Tonmi ne r a l pr obe n beschr i eben. Ange we n-
det wi r d Ti - Ka St r ahl ung, u n d u m di e Abs or pt i on di eser St r a hl ung dur c h Luf t zu mi ni mi er en,
wur de ei ne v a k u u mu n d hel i umgespf i l t e Ap p a r a t u r ent wi ckel t . Di ese k a n n mi t ander en R6nt ge n-
s t r ahl ungs quel l en be nut z t wer den, wobei ei n bet rfi cht l i cher I nt ensi t f i t sgewi nn auf t r i t t - - z.B. wi rd
Co Ke - - St r a hl ung u m 125% verst~irkt. Mi t t el s unt e r gr undbe r e i ni gt e r Di a g r a mme i st ei ne besser e
ha l bqua nt i t a t i ve Abs chi i t zung der Zu s a mme n s e t z u n g von Tonmi ne r a l ge mi s c he n m6gl i ch.
Kl ei ne Unt er s chi ede in der Zu s a mme n s e t z u n g zweier Pr obe n k6nne n dur c h di e Subt r akt i on ei nes
unt e r gr undbe r e i ni ngt e n Di a g r a mms von ander en e r ka nnt wer den. Di e An we n d u n g di eser Me t h-
ode e r mr gl i c ht di e genaue Be s t i mmu n g des Pe a k ma x i mu ms von Mi ner al en der Smekt i t gr uppe.
R E S UME N: Se descr i be u n mr t o d o pa r a cal cul ar y pos t e r i or me nt e s us t r aer el f ondo de l os
d i a g r a ma s de di f r acci 6n de mue s t r a s or i e nt a da s de mi ner al es de la arcilla. Se u s a r adi aci 6n Ti - Ka,
y p a r a mi ni mi z a r la abs or ci 6n de est a r adi aci 6n por el aire, se ut i l i za s u mont a j e a vaci o y con flujo
de hel i o. Est e mont a j e puede ser us a do con ot r as f uent es de r a yos X, of r eci endo un consi der abl e
a u me n t o de i nt ensi dad, p. e. , la r adi aci 6n Co - Ks a u me n t a s u i nt ens i dad en u n 125~. Con el f ondo
el i mi nado es posi bl e me j or a r l as es t i maci ones s emi cuant i t at i vas s obr e la compos i ci 6n de l as
mezcl as de mi ner al es de la archi l l a. Pequef i as di f er enci as en compos i ci 6n de dos mue s t r a s pue de n
ser de t e r mi na da s por di f er enci a ent r e l os dos di a gr a ma s e l i mi na ndo pr evi ament e el f ondo.
Us a n d o est e m&o d o l os mf i xi mos de di f r acci 6n de l os mi ner al es des gr upo de l as es mect i t as
pue de n se de t e r mi na dos con preci si 6n.