Org. Geochem. Vol. 20, No. 6 pp. 789-795, 1993 0146-6380/93 $6.00 + 0.

00
Printed in Great Britain. All rights reserved Copyright 1993 Pergamon Press Ltd
Detecti on and evaluation of hydrocarbons in source rocks
by fluorescence microscopy
B. ALPERN, 1'2 M. J. LEMOS DE SOUSA, 2 H. J. PINHEIRO 2 and X. ZHU 3.
'23 bis, rue des Cordelitres, 75013 Paris, France, 2Organic Petrology Unit, Department of Geology, Faculty
of Sciences, University of Porto, Praga Gomes Teixeira, 4000 Porto, Portugal and 3Central Laboratory
of Petroleum Geology, P.O. Box 916, 214151 Wuxi, People's Republic of China
Abstract--In pursuing the detection of hydrocarbons in sedimentary rocks by conventional petrological
methods, an attempt has been made to correlate standard fluorescence parameters with the quality and
quantity of hydrocarbons present in crushed rocks embedded in epoxy resin. The capacity of the
embedding resin, commonly used in the preparation of petrographic samples, to extract and physically
fix hydrocarbons is recognized. This phenomenon permits one to measure monochromatic fluorescence
parameters, 1 546 and Q (u.v.) 650/500, on the trapped hydrocarbons and to correlate these parameters
with selected geochemical data.
The potential application of using these petrological parameters to evaluate oil quantities and qualities
is tested using real case studies. It is also shown that it is possible to directly detect the mature zone in
each case.
Key words----crude oil, epoxy resin, fluorescence, geochemistry, hydrocarbon, maturation, petrology,
reservoir, source rock
INTRODUCTION
Oi l is nor mal l y present in source rocks havi ng ent ered
t he mat ur at i on zone (oil window). However, it is al so
found in ot her sedi ment ary rocks due t o its mi grat i on.
I n this st udy, novel pet r ogr aphi c techniques are used
t o i dent i fy such occurrences of oil.
Fluorescence propert i es of hydr ocar bons have been
i nvest i gat ed by many aut hors, viz. Ot t enj ann (1980),
Ber t r and et al. (1986), Hagemann and Hol l er bach
(1986), Mar t i nez and Connan (1989). I t has been
est abl i shed cl earl y t hat t he fluorescence of hydr o-
car bons is rel at ed t o t he di st r i but i on of n el ect rons in
ar omat i c ri ngs and conj ugat ed doubl e bonds. Dur i ng
u.v. exci t at i on n el ect rons are exci t ed f r om one or bi t al
t o t he ot her, and upon ret urni ng t o t hei r gr ound
state, emi t phot ons ( r adi at i on) in t he visible spect rum
(Li n and Davis, 1988).
I n t he present i nvest i gat i on an at t empt has been
made t o est abl i sh rel at i ons between fluorescence
par amet er s and oil qual i t i es (in t erms of chemi cal
composi t i on) and quant i t i es, bear i ng in mi nd t hat t he
met hodol ogy empl oyed differs f r om ot hers in t hat i t
at t empt s t o pr of i t f r om t he following: (1) recogni t i on
t hat hydr ocar bons are ext ract ed and t r apped by t he
epoxy resin in whi ch rocks are embedded, (2) hydr o-
car bons become det ect abl e by fluorescence exci t at i on,
and (3) t he monochr omat i c measurement s can be
carri ed out by rout i ne met hods and equi pment similar
to t hose used for vi t ri ni t e reflectance.
*Present address: Organic Petrology Unit, Department of
Geology, Faculty of Sciences, University of Porto, Praqa
Gomes Teixeira, 4000 Porto, Portugal.
tThis epoxy resin-hardner kit is supplied by STRUERS.
Nevertheless, our concern is present l y not di rect ed
at economi c aspects of oil fields, but i nst ead at
t he relevance of mi cr o- i mpr egnat i ons or mi cr o-
neogenerat i ons det ect ed by opt i cal mi cr oscopy whi ch
may be appl i ed t o pet r ol eum prospect i ng.
METHODOLOGY
Sample preparation
Mos t of t he sampl es st udi ed were whol e r ock
cut t i ngs al r eady in a gr anul ar form. Mi l d crushi ng
( < 2 mm) was onl y necessary when t he size was t oo
bi g or i rregul ar. The cut t i ngs were washed, dri ed,
and visible cont ami nat i ons were removed. The cl ean
gr anul ar sampl es of whol e rocks were t hen embedded
in an epoxy resin ( EPOFI Xt ) and pol i shed accordi ng
t o rout i ne pet r ogr aphi c met hods (pol i shed blocks).
The epoxy resin behaves as an ext r act or of hydr o-
carbons, favouri ng a physi cal r eadj ust ment of this
fluid phase i nt o t he embeddi ng resin itself, wi t hout
a chemi cal react i on. I n addi t i on, crude oil sampl es
were mount ed on a slide (single dr op) and covered
wi t h a l amel l a (cover slip). I n some cases, each oil was
mi xed wi t h embeddi ng resin in vari ous pr opor t i ons
( 0. 5- 4. 0%) so t hat t hei r propert i es, in such an
embeddi ng medi um, coul d be studied. Wi t h reference
t o the l at t er poi nt , ot her embeddi ng resins were al so
tested and pr oduced si mi l ar results.
Hydrocarbons
On mi xi ng t he r ock sampl es wi t h t he epoxy resin,
t he hydr ocar bons which may be present in t he r ock
part i cl es are par t l y ext ract ed and easily det ect ed
under fluorescence. Since t he embeddi ng resin is very
low fluorescing ( I 546 = 6%), and once it is mi xed
OG 20/6--L 789
790 B. ALPERN e t al.
Tabl e 1. Cl assi fi cat i on o f hydr oc a r bons ( HC) present i n r ocks cr ushed a nd embedded in epoxy resin, as obser ved
under fl uorescence a n d reference to measur ement s
=utomomh
HC
dro~
f i n
wcsud~ (oonmlax)
non-ephed~ HO
mee~r ement s
O 850/ 500
not
me u u r e d
dlmmlved HC I n - r u l n i 1546 and
f o = o / 5 o o
t
wi t h a sampl e cont ai ni ng hydr ocar bons, t he resulting
fluorescence pr oper t i es are t aken t o be t hose from t he
hydr ocar bons themselves.
A por t i on of the hydrocarbons appear as automorph
hydr ocar bons, whilst anot her por t i on is dissolved
or mi xed wi t h t he embeddi ng epoxy resin (Tabl e 1).
These cat egori es have been descri bed and i l l ust rat ed
in det ai l by Al per n et al. (1992). As a br i ef summary,
aut omor ph hydr ocar bons are cat egori zed as dr ops
and films; dr ops are cl earl y spheri cal in form and
vari abl e in size. Thei r col or ranges f r om green t o
yellow-brown. Fi l ms appear as coatings of, or stuck to
particles and with varyi ng thickness. Thei r fluorescence
col or al so varies f r om green t o yel l ow-brown. On t he
ot her hand, t he di ssol ved hydr ocar bons cat egory is
compr i sed of t hose whi ch mix with t he embeddi ng
epoxy resin giving it a fluorescence col or also
dependent on t hei r chemi cal composi t i on: greenish
for sat ur at ed oils and yellow for the ar omat i c ones.
Optical analysis
The pet r ogr aphi c equi pment used for t he fluor-
escence measur ement s is compr i sed of a Lei t z mi cro-
scope ( MPV Combi ) with reflectance and fluorescence
at t achment s. Ul t r avi ol et fluorescence condi t i ons were
obt ai ned wi t h a HBO 100 W mer cur y l amp, an UGI
exci t at i on filter ( I = 365 nm), a K430 bar r i er filter
and a TK400 di chroi c mi r r or . The i nt ensi t y of the
embeddi ng epoxy resin was measured at I = 546 nm,
usi ng an uranyl -gl ass as a st andar d. The size of t he
field measur ed on t he embeddi ng resin was 100/~m 2,
and a mi ni mum of 50 measurement s were recorded
for each sample. Al t hough the r at i o of Q = 650/500
in u.v. l i ght may be measur ed bot h for t he hydr o-
car bons (films, drops), when present, and for t he
embeddi ng resin, in the present st udy it was onl y done
on t he epoxy embeddi ng resin. The above measure-
ment s were car r i ed out using a 50 x oil i mmersi on
objective, and t he spect ral fluorescence of t he free
crude oils were per f or med usi ng wat er i mmersi on.
AP P LI CATI ONS TO P ETROLEUM P ROS P ECTI NG
Correlation of Q (u.v.)650/500 with oil quality
Q (u.v.)650/500 measurement s were carri ed out on
10 selected oils which r epor t ed wide vari at i ons in
chemical composi t i on (Tabl e 2).
A good posi t i ve cor r el at i on was obt ai ned between
Q (u.v.)650/500 and the percent age of ar omat i cs
( ARO) pl us heavy pr oduct s ( HP = resins + asphal t -
enes) (Fig. 1). That is to say t hat ar omat i c oils are
mor e red shifted t han sat ur at ed oils. This was con-
firmed by spect ral fluorescence measurement s (using
a wat er i mmersi on objective) t hat showed ar omat i c
oils had 2max values at l onger wavelengths t han
sat ur at ed oils (e.g., Fi g. 2).
Since our r out i ne met hods are carri ed out on r ock
sampl es embedded in epoxy resin, t he crude oils were
mi xed wi t h resin and fluorescence measurement s were
performed. The resul t i ng cor r el at i on between the
col or, expressed by Q, and qual i t y, as SAT/ ( ARO
+ HP), r emai n excellent wi t h r = - 0 . 9 5 7 (Fig. 3). I t
shoul d be not ed t hat , for each crude oil, t he Q values
obt ai ned on each of t he different concent r at i ons of
oil mi xed with the epoxy resin, were so similar t hat
t he average Q value was consi dered representative.
Thi s l ast poi nt was a val i d ar gument in f avour of
Tabl e 2. The genet i c nat ur e a nd chemi cal compos i t i on of t en
cr ude oils
Resi ns +
Genet i c Sat ur at ed Ar oma t i c asphal t enes
Oil No. char act er ( %) ( %) ( %) Densi t y
1 Mar i ne 82.0 14.0 4. 0 0.81
2 Cont i nent al 84.9 12.9 2.2 0.87
3 Mar i ne 26.7 56.3 16.9 0.97
4 Mar i ne 58.8 29.0 10.8 0.93
5 Mar i ne 88.5 l 1.5 - - 0.79
6 Mar i ne 63. 0 33.1 3.9 0.81
7 Mar i ne 42.0 27.0 31.0 0. 90
8 Mar i ne 28.0 47. 0 25.0 0.87
9 Mar i ne 63.3 29.3 7.4 0.80
l 0 Mar i ne 60. 0 22.0 18.0 0.83
Hydrocarbon detection and evaluation by fluorescence microscopy 791
1.0 - - Ln (O) =- 5. 53 + 1.242" Ln (ARO+HP) /
0.8 - - N= 10, r =0. 951 + 7 ~ Aromatic
O
0,6 - -
u-~c~ 9 ~ + 8
0.4 -
+
>
+10
+ ~ e d k 6
0.2 - +2
I t I I
0 20 40 60 80 100
ARO+HP ( %)
Fi g. 1. Cor r el at i on bet ween opt i cal par amet er Q ( UV) on free oil (slide) and oil qual i t y ( ARO + HP %)
for 10 sel ect ed oils.
100
50
0 I I I I I I I ! I I I I I I I
400 500
100
50
Wavel engt h ( nm)
I I ! I I | I I I
600 700
o . . ,
/
0
400
Cr ude oi l No 5
SATURATED
~.MAX Q 650 SAT/
/500 (ARO+HP)
ZU12 440 0.05 7.62
ZUI3 440 0.04 7.62
ZU14 440 0.04 7.62
ZU15 440 0.03 7.62
ZU16 430 0.04 7.62
Mean 438 0. 04 7. 62
[
' ~ " ZU12 ZU13 - - ZU14
ZU15 ZU16 Mean
f 'X
I ! I ! I I I I ! I ! ! I I I I I I
500 600 700
Wavel engt h ( nm)
Fi g. 2. Spect ral fl uorescence o f No. 5 and No, 3 free oil sampl es.
Cr ude oil No 3
AROMATI C
k MAX Q 650 SAT/
/500 (ARO+HP)
ZUI 560 1,06 0.36
ZU2 580 0.99 0.36
ZU3 560 0.96 0.36
ZU4 580 1.02 0.36
ZU5 580 1.03 0.36
Mean 572 1.01 0,36
- - - ~ ZU1 ZU2 ZU3 [
I
........ ZU4 ..... ZU6 Mean
792 B. ALPERN et al.
1. 0 m
Q ( o i l i n r e s i n ) = 0 . 2 4 4 - 0 . 0 3 1 ' ( S A T / ( A R O + H P ) )
N = 10, r = - 0 . 9 5 7
..= 0.8 -
r e d
& 0 . 6 -
~ 0 . 4 - -
A r o ma t i c
> 3 7
+ 4
~-, I ~ - ~ 6
0 " 2 1 8 ~ 10 "~ '9 ~ 1 . . . . . . . . . . . . ~
~ ~ 2 S a t u r a t e d
| b Z u e I I I " - - ' ~ , I I
0 2 4 6 8 10
SAT/(ARO+HP)
Fig. 3. Correlation between Q (UV) of embedding epoxy resin mixed with 1% oil (polished block) and
oil quality [SAT/(ARO + HP)].
considering Q as an optical paramet er grossly related
t o oil quality (Alpern et al., 1992).
Correlation o f 1546 with oil quantity
As previously mentioned the fluorescence properties
of oils are predomi nant l y related t o aromat i c struc-
tures. Therefore, onl y when the quality, i.e., relative
percentage of aromat i c hydrocarbons in an oil, is
known (Q value) can the fluorescence intensity ( I 546)
be related t o oil quantity. This fact was established
from the samples prepared by mixing increasing
quantities of each oil (0. 5-4. 0%) with epoxy resin and
measuri ng their respective fluorescence intensities.
I t is clear f r om Fig. 4 t hat if the quality is fixed,
for example Q (u.v.) equal t o 0.4, the 1 546 value is
quant i t y dependant. It may effectively be observed
t hat 1 546 increases from 200 t o 450 with an increase
in oil pr opor t i on (concentration) from 0.5 t o 4.0%.
Since it is suspected t hat bot h 1546 and Q 650/500
may simultaneously influence quality and quantity,
future investigations are presently being directed at
establishing the mathematical correlations between
these four parameters.
Estimation o f oil quantities in petroleum boreholes
The most frequent parameters used t o estimate
oil quantities in borehole are the weight percent
of extractable organic mat t er and the S1 yield from
Rock Eval. However, we believe t hat the extraction
capacity of the embedding epoxy resin is equal or
may be even greater t han t hat of the above mentioned
parameters, because in previously extracted samples,
the fluorescence intensity ( I 546) of the embedding
resin can still remain significant. The reasons for this
are t hought to be related with the following three
aspects (Table 3):
4 5 0
4 0 0 - - 4 . 0
/4o / J
3 5 0 - -
~- 3 0 0 - , , , / 3 " 0 v / ' ~ i / /
.--. / / / 1 . 0 % 0 . 5 %
2 o_
2 0 0 %: N = 7, r = 0 . 9 0 7 , 1 = - 2 9 . 3 7 + 5 6 8 . 2 6 * Q
1 5 0 / / / / / 10~: ~= 7, r = 0 9 0 0 , , = - 3 1 4 2 + 7 0 6 4 6 " 0
~'~ / / / / / 2 . o % : ~ = 7 , r = 0 . 9 0 2 , , - 4 6 . 9 1 .,- 9 7 6 . 0 8 " 0
~ 1 0 0 ///// 3 . 0 ~ : ~ = 7, r = 0 9 1 5 , ~ = - 4 4 7 8 + 1 1 O8 3 5 " 0
- 5 o / / / / j r=0938,,=-4009+121545.0
/ f / I I I I I I I I I
0 0. 1 0 . 2 0. 3 0 . 4 0. 5 0 . 6 0 . 7 0. 8 0 . 9 1. 0
Q ( U V ) [ o i l i n r e s i n ]
Fig. 4. Correlation between optical parameters [Q (UV) and I 546 (UV)], oil qualities and increasing
percentages (0.5~.0%) of oil mixed with embedding epoxy resin (polished blocks). If the quality is fixed,
then 1 546 is proportional to hydrocarbon percent.
Hy d r o c a r b o n d e t e c t i o n a n d e v a l u a t i o n b y f l uor e s c e nc e mi c r o s c o p y
Tabl e 3. Compar i son between the ext ract i on capaci t y of the embeddi ng epoxy resin and geochemical t ool s (after
Espitali6 e t a l . , 1985, modified). The shaded area corresponds t o the HC opt i cal l y detected
793
EXl l M~AI KJE
OI NI AI ~ NATlrlER
IqOGK EVAL
PAIqAMETIIm
CHLOftOFOlm EXTRACT
(mnJulm)
EXl l U, ~l ON BY
EMIDEDONm RESIN
HY DROCARBONO HEAWCOMPQUNDe
Ca r b o n n u mb e r
, i I ~ I g i * i J , , i ~ ' I ' t ,
, i c , o & *
I '
8 O 8 ' 1 1
j ~
i
Re u l n e
+
Ao p h ml t e n e a
I-
i
~ l a o l o
(1) I t is known t hat the SI yield does not corre-
spond t o the total free hydr ocar bons in rocks,
but onl y t o the volatile fraction, typically
cont ai ni ng < 33 carbons.
(2) Gas bubbles, perhaps belonging t o SO, are often
included in oil drops visibly t rapped in the
embedding epoxy resin.
(3) The fact t hat the C33+ fraction is included
in the $2 yield and not in the SI yield, explains
the existing correlations between 1 546 and $2
(Fig. 6). This is so because the embedding
epoxy resin is capable of separating and fixing
at least a part of the C33 + fraction.
I n testing the application of the routine fluorescence
parameters, three real case studies are presented. I n
borehole CAI (Fig. 5) the correlation between 1 546
and S1 is excellent with r = 0.934. I n borehole TC
(Fig. 6) the correlation is a little lower, r = 0.89, but
the variations with depth are very similar between I
546 and the Rock Eval yield data. I n the Paris basin
(Fig. 7), Toarci an oil shales are clearly detected
and the correlation between 1 546 and selected geo-
chemical dat a is again very high (r -- 0.954). The
above examples infer t hat within a given section
the oil quality seems t o be relatively const ant since the
correlations between ! 546 and geochemical dat a are
rather good.
100 - -
~ 80 - -
60 - -
,.-, 40
>
"-" 20
~ D
Ln 1 546 = 2. 57 + 0. 49* Ln $1
N = 12, r = 0. 934
,+
/
i I I i I
0 10 20 30 40 50
S1 ( % wt )
Fi g. 5. Bo r e h o l c C A 1: c o r r e l a t i o n b e t we e n 1 5 4 6 o f t h e wh o l e r o c k ( WR) e mb e d d i n g e p o x y r e s i n ( p o l i s h e d
b l o c k ) a n d S1 f r o m R o c k Eva l .
e ~
&
1 0 0 0 - -
S1 $2 $1 + $2 I 546 (UV)
1500 --
2000 --
2500
3000
3500
794 B. ALPERN e t al .
I I I i J
10 20 30 40 50
S(x) (mg/g) val ues from Rock Eval and
fl uorescence intensity (1 546 (UV) (%))
Fig. 6. Borehole TC: an example of the relation between selected Rock Eval geochemical data (S1 and
$2) and embedding epoxy resin fluorescence intensity [I 546 (UV)]. In this case the correlation coefficient
between 1 546 and S1 is 0.89.
CONCLUSIONS
Li qui d hydr oc a r bons occur f r equent l y i n sedi-
me nt a r y rocks. They ma y be t r apped and fixed i n t he
e mbe ddi ng resi n dur i ng r out i ne pet r ogr aphi c sampl e
pr epar at i on. These hydr oc a r bons ma y be r eveal ed
by f l uor escence exci t at i on o f pol i s hed resi n pl ugs,
a nd char act er i zed by t wo si mpl e f l uor escence mo n o -
c hr oma t i c par amet er s [I 546 and Q (u. v. ) 650 nm/
500 nm]. They pr ovi de i nf or ma t i on wi t h respect t o
t he oi l quant i t i es a nd qual i t i es (in t er ms o f chemi cal
c ompos i t i on) o f t he r ock. Q, expr essi ng t he fl uor-
escence col or , is mor e r el at ed t o t he oi l qual i t y and
t he cor r el at i on wi t h t he pr opor t i on o f ar omat i c hydr o-
car bons is good. I n bor ehol e case st udi es 1 546 shows
al so good cor r el at i ons wi t h S1 yi el d f r om Ro c k Eval .
The si mpl e pr ocedur es f ol l owed i n our i nvest i ga-
t i on not onl y per mi t eval uat i ons o f oi l qual i t i es and
quant i fi es, but can also det ect t her mal l y mat ur e zones.
Thus, t he hydr oc a r bon f l uor escence pr oper t i es o f
epoxy e mbe dde d r ocks pr ovi de an al t er nat i ve mat ur -
at i on pa r a me t e r whi ch suppl ement s convent i onal in-
di r ect me t hods such as spor e col or at i on, vi t r i ni t e
refl ect i vi t y, The r ma l Al t er at i on I ndex ( TAI ) , etc.
Al t hough our i nvest i gat i on pr ovi des a di r ect
me t hod of det ect i ng mat ur e zones it is still necessar y
t o di st i ngui sh sour ce r ocks f r om reservoi rs. The l at t er
is accompl i s hed by car ef ul mi cr os copi c obser vat i ons.
I 546 (UV) [embedding resin]
40 50 60 70 80 90
2000 I I I I I
2200
Toarcian 2383
2426 "~ 2400
Pliensbachian 2566
&
Sinemurian 2573
r~
2600
Hettangian 2738
2800
100
. . . .
_ 1 546 S1/TOC I 546 = 19.65 + 49.95* (S1/TOC)
N=8, r =0.954
3000 I I I
0.3 0.6 0.9 1.2 1.5
s1 ( r ag/ g) / TOC (%wt)
Fig. 7. Variations with depth and correlation between optical [I 546 (UV)] and geochemical parameters
(S1/TOC) for the Paris Basin.
Hydrocarbon detection and evaluation by fluorescence microscopy 795
Acknowledgements--The authors are indebted to Chevron
Overseas Petroleum Inc., Total, Institut Franqais du Pttrole
- - I FP, and Gabinete para a Pesquisa e Explora~fio de
Petrbleo---GPEP for supplying borehole cuttings, oils and
geochemical data, and for permission to publish corre-
sponding information. In particular, Mr G. Demaison and
Miss M. R. Cassa (Chevron), Mr J.-L. Oudin (Total), Mssrs
B. Durand and J. Espitali6 (IFP), and Mssrs J. Agnelo
Fernandes, R. Vieira, and F. Laima (GPEP) are acknowl-
edged for the assistance provided during this investigation.
Lastly, we are most grateful to Mrs M. Marques for her
support and advice on computer graphic design.
REFERENCES
Alpern B., Lemos de Sousa M. J., Pinheiro H. J. and
Zhu X. (1992) Optical morphology of hydrocarbons
and oil progenitors in sedimentary rocks--relations with
geochemical parameters. Publ. Mus. Labor. miner, geol.
Fac. Ci~nc. Porto N.S., 3, 1-21, Pls 1-15.
Bertrand P., Pittion J.-L. and Bernaud C. (1986) Fluor-
escence of sedimentary organic matter in relation to its
chemical composition. In Advances in Organic Geochem-
istry 1985 (Edited by Leythaeuser D. and Rullk6tter J.).
Org. Geochem. 10, 641-647.
Espitali6 J., Deroo G. and Marquis F. (1985) Rock Eval
pyrolysis and its applications. Institut Fr anci s du Pttrole.
(Report ref. 33578.)
Hagemann H. V. and Hollerbach A. (1986) The fluor-
escence behaviour of crude oils with respect to their
thermal maturation and degradation. In Advances in
Organic Geochemistry 1985 (Edited by Leythaeuser D.
and Rullk6tter J.). Org. Geochem. 10, 473-480.
Lin R. and Davis A. (1988) A fluorogeochemical model
for coal macerals. Org. Geochem. 12, 363-374.
Martinez L. and Connan J. (1989) Approche de la migration
primaire des hydrocarbures par des 6tudes int6gr6es de
g6ochimie et de p6trologie organique sur roches m6res
chauf6es in vitro. In P. Bertrand, Organisat., Colloq.,
Applications de la P6trologie Organique en G6ologie,
Orl6ans, 1988. Bull. Soc. gdol. Ft. 8 S6r., 5, 5: 937-
950.
Ottenjann K. (1980) Spektrale Fluoreszenz-Mikrophoto-
metric yon Kohlen und Olschiefern. Leitz-Mitt. Wiss.
Techn. Wetzlar VIII 8, 262-272.