Copper and copper(II) porphyrins of the Cretaceous–Tertiary boundary at Stevns Klint (Denmark

Copper and copper(II) porphyrins of the Cretaceous^Tertiaryboundary at Stevns Klint (Denmark)

Pavle I. Premovic a;*, Nikola D. Nikolic a, Ivana R. Tonsa a,Mirjana S. Pavlovic b, Miroslav P. Premovic a, Dejan T. Dulanovic a

a Laboratory for Geochemisty and Cosmochemistry, Department of Chemistry, Faculty of Science, University of Nis, P.O. Box 91,18000 Nis, Yugoslavia

b Institute of Nuclear Sciences Vinca, P.O. Box 522, 11001 Belgrade, Yugoslavia

Received 10 February 1999; received in revised form 21 September 1999; accepted 28 January 2000

Abstract

High levels of copper(II) (Cu2�) were found with the major part (s 90%) of the total Cu located in the smectite (Cu:175 ppm) and kerogen (Cu: up to 1000 ppm) of the basal black marl of the Cretaceous^Tertiary (KT) boundaryinformal type sedimentary rock: the Fish Clay at Stevns Klint, Denmark. Anomalous abundance (4000 ppm) of thekerogen Cu2�-porphyrins in this marl was detected by electron spin resonance. A model is proposed in which theenormous acid rains (caused by the KT asteroid impact) washed out the humics (already enriched with Cu2�/Cu2�-porphyrins) of the top horizon of the nearby oxic soil into the Fish Clay Basin during the KT event. ß 2000 ElsevierScience B.V. All rights reserved.

Keywords: copper; porphyrins; K^T boundary; electron paramagnetic resonance; smectite; kerogen; acid rain

1. Introduction

During the debate about the Stevns Klint Cre-taceous^Tertiary (KT) boundary deposition, sev-eral genetic models were discussed in order to ex-plain the source and enrichment processes for thebase metals present in the boundary layer. Chris-tensen et al. [1] attributed the metal enrichment inthe Fish Clay at Stevns Klint (hereinafter referredto as the Fish Clay) to an accumulation of mainlyterrigenous material with minor amounts of clay

minerals of diagenetic origin. These authors sub-divided this boundary rock into four beds fromthe bottom bed II (Maastrichtian grey marl) tothe top bed V (Danian Cerithium limestone)with the basal (Maastrichtian) bryozoan chalkas the bed I.

In a benchmark paper, Alvarez et al. [2] attrib-uted the anomalous concentrations of Ir (and oth-er so-called meteoritic/partly meteoritic metals) inthe Fish Clay to global fall-out of extraterrestrialmaterials. According to these authors, this sedi-mentary accumulation was produced by the im-pact event and re£ects an intimate mixture of themeteoritic and impact ejecta. In the following10 years, such Ir anomalies have been detectedalmost all over the world, in both hemispheres

0012-821X / 00 / $ ^ see front matter ß 2000 Elsevier Science B.V. All rights reserved.PII: S 0 0 1 2 - 8 2 1 X ( 0 0 ) 0 0 0 3 0 - 3

* Corresponding author. Fax: +381-18-46-460;E-mail: [email protected]

EPSL 5390 23-3-00

Earth and Planetary Science Letters 177 (2000) 105^118

www.elsevier.com/locate/epsl

and in marine and continental sections [3]. Kyte etal. [4] suggested that only beds III and IV (here-inafter referred as III and IV) of the Fish Claycan be used to estimate the primary asteroidalfallout. As an alternative to the extraterrestrialhypothesis, a volcanic model was proposed forthe Ir abundance in the Fish Clay beds [5].None of the above theories, however, explain allof the observed features of the Danish KT bound-ary rocks.

Geochemical analyses of the Fish Clay bedsrevealed that not only Ir (and other meteoritic/partly meteoritic metals), but also non-meteoritic(terrestrial) trace metals (such as Cu and V) arealso present in these materials [1,4,6]. Whilst con-siderable attention has been paid to meteoriticmetals present in the Fish Clay, less regard hasbeen given to terrestrial metals (e.g. Cu).

The Cu concentration in the Earth's crustranges from 24 to 55 ppm, in soils from 20 to30 ppm and in carbonaceous sedimentary rocksfrom 20 to 200 ppm. In the Fish Clay, the Cucontent ranges from 50 to 80 ppm [1,4,7] and,thus, is of the same order of magnitude as inother carbonaceous sedimentary rocks [8,9]. Sedi-mentary Cu is usually associated with organicmatter, Fe/Mn oxides, clays and other minerals[8]. The distribution and chemical nature of Cuin III were thus studied with the object of gainingsome further insight into the geochemistry of Cuduring the III forming process. Interest was par-ticularly centered on terrestrial sources of Cu.

The foundation of modern organic geochemis-try is associated with Treibs' suggestion that sedi-mentary extractable (alkyl) metalloporphyrins areprincipally the diagenetic end products of chloro-phylls and bacterio-chlorophylls [10]. Nickel(II)(Ni2�) and vanadyl (VO2�) alkyl porphyrins arethe two principal types of metalloporphyrinsfound in carbonaceous sedimentary rocks. Recentelectron spin resonance (ESR) work in our labo-ratory has shown that kerogens isolated from an-cient carbonaceous sedimentary rocks of marineorigin also contain signi¢cant amounts of VO2�

porphyrins (VO2�-P) incorporated into the kero-gen structure. It is generally agreed that, the in-corporation of the porphyrin nuclei into the ker-ogen matrix is essentially due to abiotic,

diagenetic reactions and modi¢cations of initialhumic substances and associated chlorophylls[11].

Although the widespread occurrence of VO2�-Pand Ni2�-P in ancient sedimentary rocks fromdi¡erent depositional environments has been re-ported, there are few reports of porphyrins com-plexed to Cu2�. The occurrence of extractableCu2�-P is perhaps best authenticated by Palmerand Baker [12] and Baker and Louda [13,14] whoisolated such species in very low concentrations(6 1 ppm of the total organic C) from a numberof immature sedimentary rocks, giving evidencebased on accurate mass spectrometric measure-ments. Palmer and Baker [12] and Baker andLouda [13,14] proposed that these Cu2�-P mayderive from, and be markers for, oxidized terres-trial organic matter redeposited in a marine envi-ronment.

Previous geochemical studies of humic substan-ces (i.e. humic and fulvic acids) of soils (includingpeat soils) have shown large di¡erences betweenhumic fractions; though these substances cannotbe regarded as distinctly di¡erent, but merely aspart of a continuum of compounds varying inboth polarity and molecular weight [15]. On theother hand, despite the fact that the fulvic acids ofsoils contained a signi¢cant percentage of the to-tal Cu, an extremely weak ESR signal ascribed toCu2� was observed. This signal was attributed toCu2�-non-porphyrin complexes in which Cu2� iscoordinated with the oxygen atoms of the fulviccarboxylic groups. Cu2�-P are found in (insolu-ble-acid hydrolyzed) humic acid fractions in awide variety of organic-rich soils: recent soils (in-cluding peat soils) [16^18] and paleosols [19]. Ithas been suggested that these humic Cu2�-P arelargely derived from the plant chlorophylls of soilunder oxidizing conditions.

We also report here an anomalous high concen-tration of Cu2�-P (4000 ppm: determined byESR) incorporated into the kerogen structure ofIII. The Nye KlÖv and Dania boundaries in theJylland sequences near Stevns Klint were, also,investigated by ESR spectroscopy for the presenceof Cu2�-P in kerogen. These boundary layers aremineralogically identical to that at Stevns Klint[20].

EPSL 5390 23-3-00

P.I. Premovic et al. / Earth and Planetary Science Letters 177 (2000) 105^118106

2. Experimental

The experimental techniques/methods have al-ready been described earlier [6]. ESR experiments(at room temperature) were performed using Var-ian X-band [6] and Q-band [21] spectrometers.

Hyper¢ne splitting constants and g-values forall X- and Q-band spectra were measured fromline positions and corrected for second-order ef-fects. Re¢nements of these values were accom-plished by computer simulation of the spectra us-ing a program for spin quantum number s = 1/2systems.

3. Geochemical association of Cu

To obtain an indication about the chemicalnature of the Cu present in III, this boundarymaterial was analyzed for Cu by both atomic ab-sorption spectrometry (AAS) and emission spec-trometry at various stages of demineralization.The results are given in Table 1. It is obviousthat Cu occurs in various forms, including ad-sorbed on the smectite (s 60% of the total Cu)and organically bound within the kerogen (ca.30% of the total Cu). Finally, the ascorbic acid+H2O2 test [22] indicates that a very low concen-tration (10 ppm) of Cu is resident in the sul¢defraction of III. For comparison, we list in Table 2the Cu contents of smectite of other four beds ofthe Fish Clay and of the KT boundary rocks fromother than the Stevns Klint locality: for the NyeKlÖv/Dania sites. The Cu contents of the NyeKlÖv/Dania kerogens could not be estimated,

however, due to the small size of the kerogensamples. The Cu contents of the III/IV kerogenswere also determined by instrumental neutron ac-tivation analysis (INAA) and the results are sum-marized in Table 3. Although the INAA data ofCu seem to be systematically lower than the AASdata [23] both values overlap considering the an-alytical uncertainties. Schmitz et al. [24] also an-alyzed the Fish Clay beds for Cu and their Cuvalues are consistent with our values. Besidesthe Fish Clay the other two most Ir-rich marineKT boundaries are found at Caravaca (Spain)and at Woodside Creek (New Zealand). It is ofinterest that the clay/organic fractions of the bedsof these two boundaries contain enhanced concen-trations of Cu (up to 310 ppm) [7].

The X-band (Fig. 1a) and Q-band (Fig. 1c)ESR spectra of the III kerogen are typical ofCu2�-P. The X-band spectrum resembles that ofthe Cu2�-P associated with the above mentionedorganic-rich recent soils [16^18] and paleosols[19]. The ESR spectrum of Cu2�-P in the III ker-ogen (Fig. 1a) consists of four lines which are al-most equally spaced, but not of equal width.These lines are attributed to the Cu nucleus,which has a nuclear spin of 3/2. Superimposedon each Cu hyper¢ne absorption are nine absorp-tions due to the four N nuclei, each of spin 1 onthe narrowest Cu hyper¢ne line. The spacing ofthe N hyper¢ne structure is 1.45 mT. The line-widths of the Cu hyper¢ne lines increase withmagnetic ¢eld from 8 mT (for the nuclear quan-tum number mI =33/2) up to 13 mT (for mI = 1/2). The asymmetry of the Cu hyper¢ne structureis, partly, attributed to the contribution of the

Table 1Geochemical data for the Danish KT boundaries: III

Fraction Sediment ( þ 5 wt%) Cua ( þ 25 ppm) Total Cu ( þ 5 wt%)

Carbonate 6 50 30 6 20Sul¢de 5 10 6 1Smectite 30 175 s 60Kerogen* 3 800 30Sediment 100 90b 80c 100

*Elemental analysis (%): C: 63.0, H: 5.1, N: 1.4, S: 1.6 and O: ca. 29 (by di¡erence); H/C: 1.0, O/C: s 0.3 and N/C: 0.2.aThe Cu content determined by AAS.bThe Cu content obtained by summation of the fraction Cu concentrations, determined by AAS.cThe Cu content determined by emission spectrography.

EPSL 5390 23-3-00

P.I. Premovic et al. / Earth and Planetary Science Letters 177 (2000) 105^118 107

anisotropic nuclear hyper¢ne interactions to thelinewidth [25]. Small-scale ESR experiments onthe IV/Nye KlÖv kerogens gave similar but notidentical results. However, ESR analysis of theDania kerogen shows no evidence for Cu2�-Pabove the limit of detection of 20 ppm.

Simulated powder ESR spectra of the Cu2�-P(Fig. 1b,d) were computed to include two di¡erenthyper¢ne interactions. The ¢rst- and second-orderperturbation theories were used for the metal (Cu)and ligand (N) nuclear hyper¢ne interactions, re-spectively. The simulated spectra were comparedwith the experimental ones in order to estimatethe anisotropic hyper¢ne parameters. The aniso-tropic hyper¢ne parameters (which were adjustedto give the best ¢t of the experimental spectra,Fig. 1a,c) are shown in Fig. 1. We were unableto ¢nd evidence for the departure from axial sym-metry in either the experimental or simulatedspectra.

4. Boundary seawater and Cu2+

Of importance is that the syngenetic carbonatefraction of III contains a very low concentration(6 30 ppm) of Cu (Table 1). Microscopic inspec-tion of a thin section of III indicates that it con-tains predominantly calcite (CaCO3) derived ex-clusively from calcerous algal plates (coccolithss 95%). The Cu2� ion is characterized by an ionicradius of 1.00 Aî [26]. Thus, Cu2� may substitutefor the Ca2� ion in the calcite matrix since bothhave the same charge [27,28]. This would involvethe direct incorporation of Cu2� from the DanishKT seawater into the III calcite or metabolic up-take by the coccolith organisms. The low Cu con-centration of III (6 30 ppm) in the biogenic cal-cite phase, which is comparable with those in thebeds II/V [22], indicates that Cu2� was not presentin enhanced concentration in the KT seawater ofthe Fish Clay Basin. On the other hand, the con-

Table 3Geochemical data for the Danish KT boundaries: III/IV

Sediment( þ 0.5 wt%)

Cu( þ 50 ppm)a

Cu2�-P( þ 100 ppm)b

Cu ( þ 50 ppm)as Cu2�-P

Fraction (wt%)as Cu2�-P

Kerogen III 3.0 600 4000 550 s 90Kerogen IV 6 0.5 200 s 1000 ^ âThe Cu content determined by INAA.bThe Cu2�-P content determined by ESR [23]: the Cu2�-P content calculated from the Cu2� concentration using 440 as the aver-age porphyrin molecular weight [14].

Table 2Geochemical data for the Danish KT boundaries: kerogen and smectite of the KT boundary layers

The Fish Clay beds Geo-chronology

Lithology Thickness(cm)

Kerogen(wt%)

Cu*(ppm)

V*a

(ppm)Cr*a

(ppm)Ir*b

(ppb)

V Danian chalk ^ n.a. 55 ^ 145 n.a.IV Danian marl 6.0 0.5** 75 170 280 54III Danian marl 3.0 3.0** 175 195 380 77II Danian marl 1.0 6 0.2 35 80 280 10I Maastrichtian bryozoan chalk ^ n.a. 35 ^ 135 ^Nye KlÖv* Danian marl ^ 6 0.1** 6 50 180 470 ^Dania Danian marl ^ 6 0.1 6 50 200 190 ^

*In the smectite fraction.**Kerogen enriched with Cu2�-P.n.a., not analyzed.There is a systematic uncertainty of 10^20% for each metal.aPremovic et al. [6].bElliot [50].

EPSL 5390 23-3-00


centration of Cu is ca. 3U106 times (in the IIIkerogen) and 6U105 times (in the III smectite)its concentration (0.3 ppb: mainly as Cu2�) inaverage seawater [29].

5. Boundary kerogen

According to Hansen et al. [30] and Schimtz etal. [24] the III kerogen mainly represents the re-mains of freshwater algae (Botryococcus), brack-ish dynocysts and marine green algal group ofprasinophytes. The dominant phase of the III ker-ogen concentrate is organic matter: of the 3% III(Tables 1^3), which survived laboratory deminer-alization procedure, about 97% consists of organ-ic matter. Elemental analysis (Table 1) shows thatthe III kerogen has a low atomic H/C ratio (1.0)and a high O/C ratio (s 0.3) and these values arecharacteristic of a transitional type IIÎII kerogen.A well-preserved marine kerogen typically has arelatively high atomic H/C ratio (v1.2) and a lowatomic O/C ratio (6 0.1) [31]. The low H/C andhigh O/C values obtained for the III kerogen inferdilution of the marine kerogen (derived fromwater-column photosynthate) by terrestrial kero-gen (derived from terrigenous humics with lowH/C). As a rule, the terrigenous component of thetransitional type of kerogen is derived essentiallyfrom terrestrial humic substances of organic-richsoils [31]. There are, however, alternative originsof this type of kerogen: it may be either re-worked/oxidized kerogen from older marine sedi-mentary rocks or kerogenous material which wasexposed to subaerial weathering (including forest¢re/biological oxidations). According to Wolbachet al. [32] the III kerogen concentrate contains: apredominantly (s 90%) highly resistant kerogenof terrestrial origin and a reactive kerogen of ma-

Fig. 1. First-derivative room-temperature anisotropic ESRspectra of Cu2�-P in the III kerogen: (a) experimental X-band; (b) simulated X-band; (c) experimental Q-band and,(d) simulated Q-band [21] spectra. Anisotropic hyper¢neCu2�-P parameters are: Ae = 19.9 þ 0.1 mT; AP 6 1.0 mT;ge = 2.189 þ 0.002; gP = 2.039 þ 0.001 and aN = 1.45 þ 0.10 mT.

Fig. 2. FTIR spectrum of the III kerogen.

EPSL 5390 23-3-00


rine origin and a minor amount (6 1%) of ele-mental carbon particles which resemble soot/char-coal from forest ¢res or marine sedimentaryrocks. In other words, the kerogen componentswithin III have more than one origin, i.e. the IIIkerogen is therefore a hybrid or polygenetic ma-terial.

The FTIR spectrum of a representative sampleof the III kerogen is presented in Fig. 2. Theassignment of the most characteristic bands inthis spectrum is based on the FTIR spectra ofcoals and kerogens [33]:

1. The broad band centered around 3400 cm31 isdue to OH groups.

2. The peaks around 2920 and 2850 cm31 are dueto aliphatic, CH, CH2 and CH3, groups.

3. The band around 1700 cm31 is due to CNO ofcarbonyl and carboxyl groups.

4. The band at 1550^1600 cm31 is characteristicfor ionized carboxyl groups (COO3).

5. The band around 1250 cm31 is characteristicfor CÔ of phenolic groups.

6. The band around 1050 cm31 is presumably dueto alcohol groups (ROH).

In conclusion, it may be stated that the FTIRspectrum of the III kerogen is similar to the spec-tra of terrestrial immature kerogens (including lig-nites and subbituminous coals) and reworked/oxi-dized kerogens [33].

6. Cu2+/Cu2+-P within the boundary kerogen

Electron microprobe spectra of the III kerogenareas which are substantially free of Si, Al, Caand Fe, always show S as the major componentand sometimes with Cu as a minor component.Electron microprobe analyses of this kerogenshowed S (up to 3%) and Cu (up to 1000 ppm).The association of Cu with kerogen in di¡erentsedimentary rock types is not a rare phenomenonand is discussed in a number of articles. For in-stance, Patterson et al. [22] showed that a portionof the Cu in the Julia Creek bituminous shale(Cretaceous, Australia) is closely associated withthe kerogen. According to these authors, the Cu is

associated mainly with sul¢des (75% of the totalCu) and to a lesser extent with kerogen (12%) andclay (10%).

The most important result of the laboratoryleaching experiment is that Cu/ Cu2�-P show astrong association with the III kerogen. The per-centage of the total Cu found in the kerogen con-centrates is greater in the kerogen-rich samples(e.g. the III kerogen versus IV kerogen, Table3). Geochemical analysis also indicates that theIII kerogen which contains a signi¢cant amountof Cu also contains a signi¢cant amount of Cu2�-P, Table 3. These observations infer that the ori-gin of the kerogen Cu/Cu2�-P is related to theorigin of the III kerogen, i.e. its humic precursors.

The data in Table 3 indicate that Cu in the IIIkerogen (600 ppm, determined by INAA) is pre-dominantly in the Cu2�-P form (550 ppm, deter-mined by ESR): it constitutes s 90% of the totalCu. Apparently, the rest of Cu is, at least, ìnvis-ible' for ESR at room temperature. The mostplausible explanation is that the ìnvisible' Cu isin a diamagnetic state, either as Cu2� in the clus-ter structures which involves strong Cu^Cu mag-netic interactions, or, less probable, as Cu1�

which has no unpaired electron. Indeed, the redoxdiagram for Cu [34] indicates that, regardless ofphysicochemical conditions (i.e. oxic versus an-oxic milieu) in natural waters, Cu2� is the onlyCu ion available for incorporation into the humicstructures and the smectite skeleton.

7. Detrital nature of the kerogen Cu2+-P

Geochemical data provide strong evidence thatIII was deposited either under anoxic conditionsor soon after deposition the conditions becameanoxic and H2S was present in both the layerand the overlying seawater [6]. In most cases,the kerogen of this rock type is thought to bederived from humic substances with suitable func-tional (including porphyrin) groups which mayform coordination complexes with Cu2� [35].However, under strong anoxic sedimentary condi-tions (such as those of the Julia Creek blackshale) most of Cu2� is precipitated as sul¢des;Cu2�-P is exceedingly labile under these condi-

EPSL 5390 23-3-00


tions [36], but they preserve porphyrins [37]. Thegeneral absence of Cu2�-P in sedimentary rocksfrom shallow marine depositional paleoenviron-ments (e.g. the Julia Creek carbonaceous rock)is consistent with such a rationalization. In otherwords, the anoxic sedimentation conditions of IIIwere rather unfavorable for the formation of thekerogen Cu2�-P. The evidence which supports theidea that the kerogen Cu2�-P were not formed insitu (i.e. within III) is: (a) an exceptionally highconcentration (4000 ppm, Table 3) of the kerogenCu2�-P within III and (b) the low concentration(10 ppm, Table 1) of Cu associated with the sul-¢de phase of III. These results suggest that thekerogen Cu2�-P are strictly detrital in character,i.e. were transported to the Fish Clay site.

8. Organic oxic soil

In an oxic sedimentary environment Cu2� isreadily chelated to humic substances (and is ad-sorbed on the clay surface), which leads to theenrichment of this ion in the sedimentary rocks(such as III) ; thus, in this type of environment,Cu2� is capable of metallating humics [35] andtheir free porphyrins [16,38,39]. If the above rea-soning is correct, we can only surmise that Cu2�

chelation by the free porphyrins within humicsubstances had to occur in an aerated aqueousenvironment (e.g. an organic-rich oxic soil) priorto entering the Fish Clay Basin. As Cu2� precip-itates as hydrous oxides under mild acid condi-tions (pHv5.1) [26] this indicates that the soilmilieu in which humic Cu2�-P were formed musthave been acidic (pH95). According to Palmerand Baker [12] Cu2�-P in marine sedimentaryrocks represent a redeposited form of eroded/oxi-dized pigments originally present in terrestrial or-ganic accumulations (e.g. peats) and were sug-gested as biomarkers for terrestrial in£ux. Inessence, porphyrins in terrestrial sedimentaryrocks are commonly complexed with Cu2� [14].It is noteworthy that our ESR measurements in-dicate that the concentration of Cu2�-P in the IIIkerogen is in the range of 4000 ppm (Table 3),which is, at least, s 4000 times higher than itsconcentration in organic matter of immature ma-

rine rocks studied by Baker et al. [12^14]. On theother hand, the amount of Cu2�-P in the III ker-ogen is comparable with the Cu2�-P contents ofhumic insoluble-acid hydrolyzed fractions of re-cent soils [16^18] and paleosols [19]. Obviously,Cu2�/Cu2�-P in the III kerogen are not homoge-neously distributed but are associated with a spe-ci¢c part of this material which originated fromits terrestrial humic precursors. Hence, we pro-pose that terrestrial humic substances played animportant role in the Cu2�/Cu2�-P transport and¢xation through the III column. Indeed, the soilhumics show a far greater a¤nity for Cu2� thanmarine humic matter [26]. In addition, Goodmanand Cheshire [16,38] have shown that the soil hu-mic substances have a relatively great capacity forthe uptake of Cu2� as Cu2�-P when they aretreated with aqueous solutions containing Cu2�

ions.Controls on accumulation of humic matter in

sediments and sedimentary rocks are imperfectlyunderstood and have been a subject of debate formany years [15]. Modern studies of sedimentarytetrapyroles indicate that the free porphyrin con-tent of sedimentary humics is very sensitive to theduration of exposure of these materials to air O2

[14]. For this reason, long residence time in apartially watered milieu of organic-rich oxic soilsmay result in relatively low ratios of free porphyr-ins to total organic matter. In fact, preferentialdegradation of chlorophylls in oxic soils beginsduring senescence or shortly after the death ofplants [14]. Therefore, a high concentration ofCu2�-P within the III kerogen structure is notexpected. However, geochemical studies of metal-loporphyrins indicate that metal ions (such asCu2�) contribute essentially to the chemical stabil-ity of the porphyrins in sedimentary environmentsthrough geologic time [14]. In consequence, theincorporation of Cu2� into the porphyrin nucleigives these pigments enough extra-stability to per-mit their survival even under unfavorable aerobicconditions of oxic soils.

Notable is the absence of V in the III kerogen,a point overlooked in the paper by Premovic et al.[6]. Under typical aerobic conditions, the moststable forms of V in aqueous solution are gener-ally vanadate ions HnVOn33

4 [6]. As a conse-

EPSL 5390 23-3-00


quence, the V species involved in adsorption pro-cesses appear to be anionic, resulting in a relativelylow a¤nity for both sedimentary humic substan-ces [40] and clay minerals [41]. Indeed, laboratoryexperiments by Shiller and Boyle [40] have shownthat complexation by humic materials is not amajor factor in determining the aquatic behaviorof the HnVOn33

4 ions. On the other hand, in an-oxic/acid media, humic substances may reduce va-nadate ions to VO2� accompanied by chelation[6]. Consequently, in an anoxic sedimentary envi-ronment, V is almost solely present as VO2� andis available for complexation with both humicsubstances [42] and clays [43]. Thus, the absenceof V in the III kerogen is consistent with the for-mation of its humic precursors (hereinafter re-ferred as the III humics) in an oxic soil.

9. Redeposition of terrestrial humics

The kerogen concentrate of the Nye KlÖvboundary rock also contains a relatively highCu2�-P (6 1000 ppm). This rock was depositedin the same marine basin, although separated byabout 300 km from the Stevns Klint location [11].This indicates that the humics (as the progenitorsof the III/Nye KlÖv kerogens) came from a com-mon source at a considerable distance and wereponded in marine Danish KT Basin. This conclu-sion is in accordance with terrestrial/detrital hy-pothesis for these organic materials.

In the light of the above, we suggest that theredeposition of humic substances of a freelydrained (oxic) soil, near the Fish Clay Basin,served as the ultimate source of the III kerogen.Supporting this view is the observation that theacid-insoluble mineral fraction of III containsmore than 20% of the sand and silt particles (in-cluding well-rounded quartz and feldspar grains)of local terrestrial origin ([1,24,44], this work). In-deed, modern studies of sedimentary rocks showthat silt in marine sedimentary rocks is mainly ofterrestrial origin and most of it is produced duringsoil formation [26]. Evidently, Cu2� of the IIIhumics must have had its source rock on landfrom the surrounding area. An exceptionallyhigh concentration of Cu (600 ppm, Table 3) in

the III kerogen implies that Cu in the source rockwas greater than normal. The leached Cu (asCu2�) of this rock reached the organic-rich oxicsoil near the Fish Clay Basin either through solu-tion, or in-transit adsorption by terrigenous hu-mic material that was carried into the soil bystreams draining the source rock area.

Hansen et al. [30] estimated an average sedi-mentation rate for the bed I (grey chalk) of ca.7U1033 cm yr31. If the Fish Clay represents asudden in£ux of ¢ne terrigenous material intothe basin, then this marl may be considered torepresent a more rapid accumulation then a nor-mal deposition of chalk. So, we can probablysafely multiply by 10 the sedimentation rate ofbed I in order to arrive at an approximate rateof accumulation of III: i.e. 7U1032 cm yr31. Ifwe conservatively assume that this rate prevailedat the Stevns Klint locality, we calculate that the(maximum thickness) 3 cm of III represents ca. 30years of sedimentation. In our opinion, even this¢gure is a considerable overestimate of the actualtime for accumulation of III within the Fish Claydeposit.

10. Impact of acid rains

Schmitz [7] suggested that the Danish boundaryrocks were probably deposited as a result of abasin wide regression. A strong case for the rela-tion between the oceanic regressive event and thepresence of Cu2�-P in marine sedimentary rockshas been previously presented by Baker andLouda [14]. However, the formation of III con-taining initially humic substances from a soil isdi¤cult to attribute to the regression at the endof the Cretaceous, as it was relatively instantane-ous (930 years) when compared to the lengthytime of the sea withdrawal. Another possibilitycould be that the KT impact event caused large,rapid (but unpredictable) £uctuations of sea-leveldepending on the nature of the impact (land,ocean, etc.). For instance, an oceanic impactcould produce waves with the initial amplitudeof several kilometers [45]. Alternatively, massivedraining of a preexisting soil by acid rains causedby either the asteroid impact or volcanic escala-

EPSL 5390 23-3-00


tions could, also, produce a high in£ux of humicmaterials into the Fish Clay Basin during orimmediately after the KT event [46]. Prinn andFegley [46] have, however, argued that the chem-ical (weathering/leaching) e¡ects of such volcan-ism would be much less severe than those ofthe impact produced nitrogen(II) oxide (NO),largely because of the slow rate of volcanic addi-tion.

It is also noteworthy here that the presence ofkerogen (enriched with Cu2�/ Cu2�-P) in the FishClay commences abruptly with III (Table 2). Thisis consistent with the abrupt input of terrestrialhumic materials into the Fish Clay Basin duringthe KT event suggested by Premovic et al. [6].These authors also proposed that during thedeposition of IV the input of terrestrial humicsubstances was considerably reduced or ceasedto exist at all. The sudden and sharp drop inthe kerogen content (Table 2), the kerogen Cu2�

content (Table 2) and the Cu2�-P content (Table3) from III to IV is in line with such a claim.These observations are very compatible with thesudden e¡ects of a large meteoritic KT impact.Indeed, all experimental studies have shown thatthe rate of metal (e.g. Ir) enrichment of the FishClay was the greatest in early stages of the rockaccumulation (i.e. during the KT event) and de-clined almost exponentially with time after theevent.

Experimental studies and observations on mod-ern soils suggest that there is a steady decrease inhumic content with the depth through the soilpro¢le. Accordingly, the top (eluvial) horizon usu-ally contains the most of humic materials within aparticular soil. Lower horizons are made mainlyfrom mineral matter and contain minor amountof humics [26]. Tentatively, we suggest that thesudden and sharp increase of the kerogen contentin the basal black marl (i.e. III) could be ex-plained by abrupt removal of humics of the tophorizon which was ¢rst washed out from thesource soil by the KT acid rains. During the laterstages of development the soil detritus from thelower horizons swept in the Fish Clay Basin. Thisultimately led to the sharp drop in the kerogencontent (i.e. its humic precursors) from III toIV. The inorganic geochemical data on the Fish

Clay can thus be reasonably well explained underthe hypothesis that this boundary partly repre-sents redeposited materials (by the impact acidrains) of the top horizon of a near-by organic-rich oxic soil.

Simple calculation based on the Cu2�-P contentof the III kerogen (Table 3) indicates that about3% of the total N (1.4%) in this material occurs asCu2�-P. The large excess of total elemental Nover that accounted for the Cu2�-P content ofthe III kerogen suggests that other metallopor-phyrins (e.g. Ni2�-P, Ga3�-P, Mn3�-P or Fe3�-P) or even free porphyrins might be also presentwithin the III kerogen structure; Ga3�-P, Mn3�-Pand Fe3�-P were identi¢ed in humic coals [14].This, however, is only a postulate and is sup-ported by circumstantial evidence though rela-tively high concentrations of Ni and Fe havebeen detected in the III kerogen concentrate [7].Wolbach et al. [47] pointed out that in III N is 20-fold enriched over the values of underlying Creta-ceous (I/II) or overlying Tertiary (IV/V) beds: theN/C ratio is 0.20 (similar to our value of 0.2,Table 1), compared with 0.03^0.05 in most ofthe Fish Clay beds. These researchers attributedsuch an enhanced abundance of N to an enhancedpreservation of nitrogen compounds augmentedby nitrosylation by HNO2 from acid rains causedby the asteroid impact.

11. Boundary smectite and Cu

The origin of the III smectite (mainly Mg-smec-tite) has been variously attributed to the altera-tion of impact glass [48], to detritus swept in dur-ing the sea-level regression [7] or to alteration ofvolcanic ash [49]. According to Schmitz [7] theabundant presence of the microfossils of dino£a-gelletes in III [23] infers a near shore origin of theIII smectite as these microplants are brackish- orfresh-water species [7]. Premovic et al. [6] sug-gested that the Danish KT boundary clay repre-sents weathered clay (along with some asteroid/local materials) that was transported to the Dan-ish boundary sites during or immediately after theKT event. These authors claimed that most ofterrestrial metals (e.g. V) associated with the

EPSL 5390 23-3-00


smectite fraction are strictly detrital in character,i.e. were transported to the Danish boundary sitesalready contained in clay.

Table 2 shows the Cu contents within the smec-tite fractions of the beds I^V of the Fish Claybecause the smectite appears to be the major car-rier of Cu (Table 1). The di¡erence in the abun-dance of Cu in the clay fractions of the Fish Claylayers have been noted before by Schmitz et al.[24]. Fig. 3 shows the distribution of the normal-ized values of the average Cu contents of thesmectite within the beds I^V. For comparison,Table 2 lists, also, the contents of: V, Cr (as atypical partly meteoritic metal) and Ir (as a typi-cal meteoritic metal). The normalized averageabundances of these three metals are also pre-sented in Fig. 3. Five generalizations can bedrawn from the data presented in Table 2 andFig. 3: (1) the modest Cu enrichment of theFish Clay smectite begins with the reduced blackmarl III, but not with the lower (underlying)layers I/II ; (2) the III smectite is only 2.5 timeshigher in Cu than the average sedimentary clay(50^70 ppm [8,9]) ; (3) the III smectite is enrichedin Cu by 5 times in comparison with smectite inthe beds I/II below III; (4) the smectite within thebeds IV/V above III always contains more Cuthan its counterpart in the layers I/II below III;and (5) analytical data for V and Cr [6] and Ir [50]show that the III smectite is enriched 2.5 times(V), s 2 times (Cr) and s 7 times (Ir) in compar-

ison with smectite of the underlying bed II. Inshort, these results indicate that Cu, V, Cr andIr show similar and moderately pronounced abun-dance peak in the smectite fraction of III.

Despite the fact that the III smectite containsmore than half of the total Cu no ESR signalcharacteristic for Cu2� within the III smectiteskeleton is observed. The most likely reason isthat almost all of Cu is in a diamagnetic state,as Cu2� in the cluster structures which involvesstrong Cu2�-Cu2� magnetic interactions.

12. Separate formation of humics and smectite

The absence of V in the III kerogen and theabundant presence of both V (195 ppm) andVO2� (215 ppm) in the III smectite [6,23] implythat the III humics and the III smectite had di¡er-ent sites of formation. Sedimentary clays origi-nated in oxic soils rarely exist in pure form;they usually have a high content of hydrous oxideprecipitates of Fe, Al and Mn linked to them.They occur in the clay-size (6 2 Wm) fractionand are usually mixed with clays. However, inthe more rigorous weathering environment (e.g.in a tropical climate of the Fish Clay Basin duringthe late Cretaceous) these oxides may often beeven more abundant than clay minerals in a sedi-mentary rock [26]. On the contrary, our minera-logical analysis indicates that the III smectite isextremely pure containing only trace amounts ofillite and non-phyllosilicate minerals strongly ar-guing for the separate sites of formation of the IIIhumics and the III smectite. Another piece of evi-dence for the separate formation sites of these twosedimentary components of the Fish Clay is thelow chemical stability of Cu2�-P in contact withthe smectite due to a facile demetallation of Cu2�-P in the presence of clays [51].

13. Separate origins of Cu2+ and VO2+:primary/secondary nature

Modern studies of the behavior of trace metalsin various environments indicate that V shows ahigh mobility in either neutral or alkaline milieu

Fig. 3. Distribution of Cu, V, Cr and Ir within the smectite(II, III, IV and V) normalized to the metal content of the IIIsmectite.

EPSL 5390 23-3-00


(as HnVOn334 ) and a medium mobility in reducing

media (as VO2�) [6]. It should be noted that thesmectites are largely derived from volcanic mate-rials (including glass) and their formation fromparent volcanoclastic materials frequently takesplace in an alkaline environment [52]. This mightespecially be true for the Fish Clay which containspyroclastic labradorite and Mg-smectite thoughtto have been produced by basaltic volcanism [6].Our recent ESR analysis indicated that 85% of thetotal V (195 ppm) in the III smectite occurs asVO2� (215 ppm) [23]. According to Premovic etal. [6] the V abundance in III may be relatedlargely to terrestrial source(s) which are crustaland mantle (volcanism, hydrothermalism). Theseauthors concluded that VO2� within the III smec-tite is of the primary origin, i.e. it was incorpo-rated into the clay skeleton during the smectitegenesis. Also signi¢cant is that the Cu2� andVO2� incorporation into the smectite structuregenerally seems to be mutually exclusive: the in-corporation of Cu2� into the smectite skeletonrequires an oxic milieu and VO2� a more reducingenvironment. Thus, the results reported herein im-ply separate origins for Cu2� and VO2� within theIII smectite. The fact that no joint occurrence ofCu2� and VO2� in the clays or the clay fractionsof sedimentary rocks has been ever reportedstrongly supports such a contention.

Unlike VO2�, Cu2� shows a medium degree of

mobility only in an oxidizing/acid environment(e.g. acid rains). Considering that the Cu2� andVO2� probably had separate origins and thatVO2� is likely to be the primary ionic constituentof the III smectite, it is almost certain that Cu2� isof the secondary nature, i.e. it was introduced intothe III smectite (formed elsewhere) rather than asyngenetic ionic entity. This incorporation prob-ably took place during the clay emplacement(from its original site to the Danish KT bounda-ries). Relatively high concentrations of Cu of thesmectite and the kerogen within III imply that thesource of Cu of this smectite and the III humicswas probably the same. It is, also, possible tospeculate that a transmetallation reaction betweenCu2�-P of redeposited humic substances and theclay was responsible for the Cu enrichment of theIII smectite. However, the fact that the kerogencontent of III is ca. 10 times lower than the smec-tite content excludes such a possibility.

Going on the assumption that the III kerogen isderived through the redeposition of humic mate-rials of an oxygenated soil into the Fish ClayBasin (as we suggested above), we o¡er the fol-lowing model (illustrated in Fig. 4) as a ¢rst ap-proximation to the processes involved.

During or immediately after the KT impactevent acid rains weathered out the smectite fromits original site (which was probably located nearthe Fish Clay Basin) on elevated land and ponded

Fig. 4. Proposed model for geochemical relations between the smectite (enriched with V), the organic-rich oxic soil and the FishClay Basin.

EPSL 5390 23-3-00


it into the basin. Simultaneously the near-by oxicsoils abundant in humics (likely in the form ofcollodial materials) enriched with Cu2�-P thenwere washed away by these rains to its presentday sedimentary locations (such as Stevns Klint/Nye KlÖv).

14. Conclusions

From the above we can draw the followingconclusions:

1. High concentrations of Cu were found in thekerogen (Cu: up to 1000 ppm) and the smectite(Cu: 175 ppm) of the basal black marl (i.e. III)of the Fish Clay at Stevns Klint, Denmark.

2. Anomalous high content (4000 ppm) of Cu2�-P was detected in the III kerogen by electronspin resonance.

3. It is suggested that the III kerogen derivedfrom the humics (already enriched with Cu2�/Cu2�-P) of an oxic soil near the Fish ClayBasin. The analysis of Cu2�-P within the IIIkerogen was a critical result prompting thisconclusion.

4. A model is proposed in which enormous acidrains (caused by the KT asteroid impact) rede-posited the humics of the top horizon of thenear-by oxic soil into the Fish Clay Basin.

5. Geochemistry of V and VO2� within the IIIsmectite and mineralogical data of this clayimply that the III humics and the III smectitehad di¡erent sites of formation.

Acknowledgements

This work is dedicated to the memory of Pro-fessor Slobodan Ristic who passed away during1993 and will be missed by all, especially thoseof us who were his students. We are thankful toboth the IPG Institution (France) and Ministry ofScience and Technology (Serbia) for the supportof this work. Funding support from le Ministe©refranc°ais de l'Education National, de l'Enseigne-ment Superieur et de la Recherche to P.I.P. for

his stay at Universite Pierre et Marie Curie (Paris)is gratefully acknowledged. We are also indebtedto the donor of the KT samples Dr. H.J. Hansen.The paper was reviewed by two anonymous jour-nal referees whose detailed comments are ac-knowledged with thanks.[A.C.]

References

[1] L. Christensen, S. Fregerslev, A. Simonsen, J. Thiede,Sedimentology and depositional environment of LowerDanian Fish Clay from Stevns Klint, Denmark, Bull.Geol. Soc. Denmark 22 (1973) 193^212.

[2] L.W. Alvarez, W. Alvarez, F. Asaro, H.V. Michel, Extra-terrestrial cause for the Cretaceous^Tertiary extinction,Science 208 (1980) 1095^1108.

[3] R.R. Brooks, Analytical chemists and dinosaurs, Can. J.Chem. 65 (1987) 1033^1041.

[4] F.T. Kyte, Z. Zhou, J.T. Wasson, Siderophile-enrichedsediments from the Cretaceous^Tertiary boundary, Na-ture 288 (1980) 651^656.

[5] F.L. Sutherland, Volcanism around K/T boundary time:its role in an impact scenario for the K/T extinctionevents, Earth Planet. Sci. Rev. 36 (1994) 1^26.

[6] P.I. Premovic, N.Z. Pavlovic, M.S. Pavlovic, N.D. Ni-kolic, Physicochemical conditions of sedimentation ofthe Fish Clay from Stevns Klint, Denmark, and its detri-tal nature: vanadium and other supportive evidence, Geo-chim. Cosmochim. Acta 57 (1993) 1433^1446.

[7] B. Schmitz, Origin of microlayering in worldwide distrib-uted Ir-rich marine Cretaceous/Tertiary boundary clays,Geology 16 (1988) 1068^1072.

[8] D.E. Baker, J.P. Senft, Copper, in: B.J. Alloway (Ed.),Heavy Metals in Soils, Blackie Academic and Professio-nal, London, 1995, pp. 179^205.

[9] R.R. Brooks, R.D. Reeves, X.H. Yang, D.E. Ryan, J.Holzbecher, J.D. Collen, V.E. Neall, J. Lee, Elementalanomalies at the Cretaceous^Tertiary boundary, Wood-side Creek, New Zeland, Science 226 (1984) 539^542.

[10] A. Treibs, Chlorophyll- und Ha«minderivate in organi-schen Mineralsto¡en, Angew. Chem. 49 (1936) 682^686.

[11] P.I. Premovic, M.S. Pavlovic, N.Z. Pavlovic, Vanadium inancient sedimentary rocks of marine origin, Geochim.Cosmochim. Acta 50 (1986) 1923^1956.

[12] S.E. Palmer, E.W. Baker, Copper porphyrins in deep-seasediments: A possible indicator of oxidized terrestrial or-ganic matter, Science 201 (1978) 49^51.

[13] E.W. Baker, J.W. Louda, Highly dealkylated copper andnickel etioporphyrins in marine sediments, in: P.A.Schenck, J.W. de Leeuw and G.W.M. Lijmbach (Eds.),Advances in Organic Geochemistry 1983, PergamonPress, Oxford, 1984, Org. Geochem. 6, 183^192.

[14] E.W. Baker, J.W. Louda, Porphyrin in the geological re-cord, in: R.B. Johns (Ed.), Biological Markers in the

EPSL 5390 23-3-00


Sedimentary Record, Elsevier, Amsterdam, 1986, pp. 125^225.

[15] F.J. Stevenson, Humus Chemistry: Genesis, Composition,Reactions, John Wiley and Sons, New York, 1982.

[16] B.A. Goodman, M.V. Cheshire, The occurrence of cop-per^porphyrin complexes in soil humic acids, J. Soil Sci.27 (1976) 337^347.

[17] M.V. Cheshire, M.I. Berrow, B.A. Goodman, C.M. Mun-die, Metal distribution and nature of some Cu, Mn and Vcomplexes in humic acid: An electron paramagnetic reso-nance study, Geochim. Cosmochim. Acta 41 (1977) 1131^1138.

[18] A.L. Abdul-Halim, J.C. Evans, C.C. Rowlands, J.H. Tho-mas, An EPR spectroscopic examination of heavy metalsin humic and fulvic acid soil fractions, Geochim. Cosmo-chim. Acta 45 (1981) 481^487.

[19] N. Senesi, G. Calderoni, Structural and chemical charac-terization of copper, iron and manganese complexesformed by paleosol humic acids, Org. Geochem. 13(1988) 1145^1152.

[20] M.R. Rampino, R.C. Reynolds, Clay mineralogy of theCretaceous^Tertiary boundary clay, Science 219 (1983)495^498.

[21] B. Clozel, Les defauts paramagnetiques induits par irra-diation dans les kaolinites, Approche experimentale et im-plications geochimiques, Ph.D. Dissertation, Paris 7 Uni-versite, 1991.

[22] J.H. Patterson, A.R. Ramsden, L.S. Dale, J.J. Fardy,Geochemistry and mineralogical residences of trace ele-ments in soil shales from Julia Creek Queensland, Austra-lia, Chem. Geol. 55 (1986) 1^16.

[23] N.D. Nikolic, Vanadium and Physicochemical Conditionsof Sedimentation, Ph.D. Dissertation, University of Nis,1999.

[24] B. Schmitz, P. Andersson, J. Dahl, Iridium, sulfur iso-topes and rare earth elements in the Cretaceous^Tertiaryboundary clay at Stevns Klint, Denmark, Geochim. Cos-mochim. Acta 52 (1988) 229^236.

[25] B.A. Goodman, J.B. Raynor, Electron spin resonance oftransition metal complexes, Adv. Inorg. Chem. Radio-chem. 13 (1970) 135^362.

[26] A. Kabata-Pendias, H. Pendias, Trace Elements in Soilsand Plants, CRC Press, Boca Raton, FL, 1984.

[27] A. Heydemann, Adsorption aus sehr verdu«nnten Kupfer-lo«sungen an reinen Tonmineralen, Geochim. Cosmochim.Acta 15 (1959) 305^329.

[28] M.L. Franklin, J.W. Morse, The interaction of manga-nese(II) with the surface of calcite in dilute solutionsand seawater, Mar. Chem. 12 (1982) 241^254.

[29] D.M. Hirst, Geochemistry of sediments from eleven BlackSea cores, in: E.T. Degens, D.A. Ross (Eds.), The BlackSea Geology, Chemistry and Biology, Memoir 20, Am.Assoc. Petr. Geol., Tulsa, OK, 1974, pp. 430^435.

[30] H.J. Hansen, R. Gwozdz, K.L. Rasmussen, High-resolu-tion trace element chemistry across the Cretaceous^Terti-ary boundary in Denmark, Rev. Esp. Paleont. 1987,21^29.

[31] B.P. Tissot, D.H. Welte, Petroleum Formation and Oc-currence, Springer Verlag, Berlin, 1984.

[32] W. Wolbach, R. Lewis, E. Anders, Cretaceous extinc-tions; evidence for wild¢res and search for meteoritic ma-terials, Science 230 (1985) 167^170.

[33] P.G. Rouxet, P.L. Robin, G. Nicaise, Characterization ofkerogens and their evolution by infrared spectroscopy, in:B. Durand (Ed.), Kerogen-Insoluble Organic Matter fromSedimentary Rocks, Editions Technip, Paris, 1980, pp.163^190.

[34] D.G. Brookins, Eh^pH Diagrams for Geochemistry,Springer-Verlag, Berlin, 1988.

[35] D.M. McKnight, G.L. Feder, E.M. Thurman, R.L. Wer-shaw, J.C. Westall, Complexation of copper by aquatichumic substances from di¡erent environments, Sci. TotalEnviron. 28 (1983) 65^76.

[36] J.W. Buchler, Static coordination chemistry of metallo-porphyrins, in: K.M. Smith (Ed.), Porphyrins and Metal-loporphyrins, Elsevier, Amsterdam, 1975, pp. 157^231.

[37] M. Brongersma-Sanders, On conditions favoring the pres-ervation of chlorophyll in marine sediments, in: Proc. 3rdWorld Petrol. Congr., The Hague, Sect. I, 1951, pp. 401^411.

[38] B.A. Goodman, M.V. Cheshire, Electron paramagneticresonance evidence that copper is complexed in humicacid by the nitrogen of porphyrin groups, Nature NewBiol. 244 (1973) 158^159.

[39] P. Hambright, Dynamic coordination chemistry of metal-loporphyrins, in: K.S. Smith (Ed.), Porphyrins and Met-alloporphyrins, Elsevier, Amsterdam, 1975, pp. 233^278.

[40] A.M. Shiller, E.A. Boyle, Dissolved vanadium in riversand estuaries, Earth Planet. Sci. Lett. 86 (1987) 214^224.

[41] B. Wehrli, Vanadium in der Hydrospha«re; Ober£a«chen-komplexe und Oxidationskinetik, Ph.D. Dissertation,ETH Zu«rich, 1987.

[42] M.D. Lewan, Geochemistry of Vanadium and Nickel inOrganic Matter of Sedimentary Rocks, Ph.D. Disserta-tion, University of Cincinnati, Cincinnati, OH, 1980.

[43] B. Wehrli, W. Stumm, Vanadyl in natural waters: adsorp-tion and hydrolysis promote oxygenation, Geochim. Cos-mochim. Acta 53 (1989) 69^77.

[44] N.J. Evans, D.C. Gregoire, W.D. Goodfellow, N. Miles,J. Veizer, The Cretaceous^Tertiary ¢reball layer, ejectalayer and coal seam: Platinium-group element contentand mineralogy of size fractions, Meteoritics 29 (1994)223^235.

[45] R.H. Dott, Episodic sedimentation: how normal is aver-age? How rare is rare? Does it matter?, J. Sediment. Pet-rol. 53 (1983) 5^23.

[46] R.G. Prinn, B. Fegley, Bolide impacts, acid rain, and bio-spheric traumas at the Cretaceous^Tertiary boundary,Earth Planet. Sci. Lett. 83 (1987) 1^15.

[47] W. Wolbach, I. Gilmour, E. Anders, C. Orth, R. Brooks,Global ¢re at the Cretaceous^Tertiary boundary, Nature334 (1988) 665^669.

[48] M. Kastner, F. Asaro, H.V. Michel, W. Alvarez, L.W.Alvarez, The precursor of the Cretaceous^Tertiary bound-

EPSL 5390 23-3-00


ary clays at Stevns Klint, Denmark, and DSDP hole 465a,Science 226 (1984) 137^143.

[49] W.C. Elliott, J.L. Aronson Jr., H.T. Millard, E. Gierlow-ski-Kordesch, The origin of the clay minerals at the Cre-taceous/Tertiary boundary in Denmark, Geol. Soc. Am.Bull. 101 (1989) 702^710.

[50] W.C. Elliot, Origin of the Mg-smectite at the Cretaceous/

Tertiary (K/T) boundary at Stevns Klint, Denmark, ClaysClay Miner. 41 (1993) 442^452.

[51] F. Bergaya, H. VanDamme, Stability of mettaloporphyr-ins adsorbed on clays. A comparative study, Geochim.Cosmochim. Acta 46 (1982) 349^360.

[52] C.E. Weaver, L.D. Pollard, The Chemistry of Clay Min-erals, Elsevier, London, 1973.

EPSL 5390 23-3-00


Documents

Copper and copper(II) porphyrins of the Cretaceous–Tertiary boundary at Stevns Klint (Denmark