Petroleum geochemistry of the deepened Lopra-1/1A re-entry well, Faroe Islands 67© GEUS, 2006. Geological Survey of Denmark and Greenland Bulletin 9, 67–77. Available at: www.geus.dk/publications/bull Petroleum geochemistry of the deepened Lopra-1/1A re-entry well, Faroe Islands Jørgen A. Bojesen-Koefoed and H. Peter Nytoft The Lopra-1/1A re-entry well was drilled as a stratigraphic test with no immediate exploration objec- tives. Hence, petroleum geochemical studies were of limited extent, and restricted to non-destructive analyses. The presence of natural petroleum hydrocarbons could not be confirmed with certainty, but hydrocarbons extracted from the hydrochloric acid solute of a calcite vug present in RSWC #1 (3543 m), may represent indigenous petroleum since hydrocarbon-bearing fluid inclusions have been re- ported from the same sample. These hydrocarbons show some similarities to petroleum generated from the Upper Jurassic – Lower Cretaceous Kimmeridge Clay type source rocks present in surround- ing areas. Except for this sample, the results generally show the presence of a variety of contaminants of different origins such as ‘naturally greasy fingers’ (squalene and cholesterol), cosmetics such as chap stick or hand lotion (e.g. esters such as butyl-stearate, stearyl-palmitate, vitamin A), plasticisers (phtha- lates), diesel oil and ‘pipe dope’. Keywords: oil traces, organic geochemistry, contamination, Faroes, North Atlantic, Lopra ______________________________________________________________________________________________________________ Geological Survey of Denmark and Greenland, Øster Voldgade 10, DK-1350 Copenhagen K, Denmark. E-mail: jbk@geus.dk As part of the preparatory activities prior to an expected future licensing round on Faroese territory, a consortium of oil companies (see Heinesen et al. 2006, this volume) undertook to drill the Lopra-1/1A re-entry well as a strati- graphic test. The primary objectives of the well were to obtain lithological and stratigraphic information on the deepest parts of the Faroese flood basalt sequence as well as its substratum, and to acquire information related to the assessment of the petroleum exploration prospectivity of the area. The original Lopra-1 well was drilled in 1981 and reached a total depth (TD) of 2178 m. The well penetrated a suc- cession of flood basalts with minor tuffs and two dolerites (probably dykes). The drilled sequence was interpreted as part of the Faroese lower basalt formation or series (Waag- stein et al. 1984; Waagstein 1988). The Lopra-1 well was re-entered in 1996 and deepened to a TD of 3565 m. The new drilling showed that the subaerial flood basalt se- quence was underlain by a subaqueous hyaloclastite-ba- salt succession, comprising the interval 2550–3565 m and consisting predominantly of lapilli-tuffs, tuff-breccias and beds of blocks of basalt. No clastic sedimentary rocks were recorded (Boldreel 2006, this volume). The deepened Lopra-1/1A re-entry well had no direct hydrocarbon exploration objectives and, as a technical consequence, other types of investigations took priority over petroleum geochemistry. This was particularly the case since no sedimentary rocks were penetrated and nei- ther were any evident shows detected during drilling. Ac- cordingly, the analytical programme carried out with re- spect to petroleum geochemistry was somewhat limited in scope and character. Previously, various indications had been recorded of the presence of hydrocarbons in the original Lopra-1 well pri- or to deepening (Jørgensen 1984; Laier et al. 1997): 1. During drilling in 1981, a gas show was noted at 2008 m bRKB and ‘wax’/bitumen was observed on zeolites. 2. In 1983, gas at a pressure of 19 bar was sampled at the wellhead. The gas flowed at approximately 1 m3 per day and consisted of methane (72%), N2 (27%), and traces of higher hydrocarbons (Jacobsen & Laier 1984). GEUS Bulletin no 9 - 7 juli.pmd 07-07-2006, 14:1967 68 In addition, water with an oil-film was produced by the well, but the oil film was not analysed in detail. Based on its isotopic composition, it was concluded that the gas was principally of thermogenic origin, al- though the gas was unusually ‘dry’ (Laier et al. 1997). 3. In 1992, oil-film on water from the well was sampled, and minor amounts of gas escaped from the wellhead. Based on geochemical analysis, it was suggested that the oil originated from a mature siliciclastic source rock, which was presumed to be present below the basaltic cover (Laier et al. 1997). 4. During a VSP-survey in 1994, a logging tool was low- ered into the well and a thick, oil-smelling mud-slurry was observed sticking to the tool and wire when re- turned to the surface. Minor amounts of gas were not- ed. Analysis of the slurry revealed the presence of petro- leum components similar to those detected in water samples from the well. Moreover, wax-coatings on zeolite minerals have been re- corded in several outcrops on the Faroe Islands (Jørgensen 1984; Laier et al. 1997). In order to check if these indications of the presence of petroleum in the Lopra-1 well prior to deepening could be further substantiated, a minor petroleum geochemical analytical programme was carried out on samples from the deepened well, the results of which are reported below. Samples and methods Since the Lopra-1/1A re-entry well was drilled as a strati- graphic test, samples were principally reserved for investi- gations directly related to the main objectives of the pro- gramme, and petroleum geochemical studies were gener- ally restricted to non-destructive analyses. Hence, sam- ples could not be crushed and analyses had to rely on or- ganic matter extracted from the surface and the directly accessible pore-spaces of the recovered rock samples. This procedure is necessarily sub-optimal, since, as will be dem- onstrated, the risk of contamination is overwhelming. However, for pre-determined reasons other types of inves- tigations had been assigned greater priority. All analyses were carried out on rotary sidewall cores (RSWC), 58 of which were collected over the section pen- etrated in the deepened well below the TD of the original Lopra-1 well, i.e. deeper than 2178 m bRKB. All samples consisted of basaltic volcanics, i.e. lavas, tuffs and hyalo- clastic breccias. Upon receipt, all RSWC samples, which came in screw- cap glass containers, were taken to a darkroom and checked by organolfactoric means for petroleum odour, and for visible fluorescence by means of a hand-held UV-lamp. Based on the presence of a distinct petroleum odour and weak to clear fluorescence, seven samples were selected for further study (Table 1). Table 1. Sample identification, solvent extraction and gas chromatographic key data Sample RSWC #51 RSWC #45 RSWC #36 RSWC #33 RSWC #24 RSWC #13 RSWC #1 RSWC #1 RSWC #1 Drilling mud Drilling mud Pipe dope a in wt% of total extract. b in wt% of maltene fraction. c pristane/phytane ratio, from gas chromatography. d carbon preference index, calculated over the interval nC22–32. Depth (m b. KB) 2361 2450 2570 2630 3076 3438 3543 3543 3543 2900 3200 Vesicular basalt Basalt Welded tuff Welded tuff Basalt Lapillic tuff Tuff w. calcite vug Tuff w. calcite vug Tuff w. calcite vug Entire RSWC immersed in DCM Entire RSWC immersed in DCM Entire RSWC immersed in DCM Entire RSWC immersed in DCM Entire RSWC immersed in DCM Entire RSWC immersed in DCM Entire RSWC immersed in DCM Calcite vug dissolved in HCl, organic extract recovered by shaking with DCM Tuff chip, counterpart of calcite vug, crushed and extracted Approximately 10 g of drilling mud before addition of diesel, rinsed by DCM Approximately 10 g of drilling mud after addition of diesel, rinsed by DCM Anti-seize compound for drilling rods sample supplied by DANOP Extract recovery (mg) 1.0 0.8 0.6 0.8 0.7 1.1 n.d. 1.1 6.3 5.7 23.9 n.a. Asphalt- enesa n.a. n.a. n.a. n.a. n.a. n.a. n.a. 27.3 76.2 7.0 2.5 2.1 Satur- atesb n.a. n.a n.a. n.a. n.a. n.a. n.a. n.a. 22.2 67.3 68.0 57.8 Aro- maticsb n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. 11.1 19.2 22.2 24.5 NSOb n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. 66.7 13.5 9.9 17.7 pr/phc 1.61 1.50 1.52 1.41 1.36 1.34 1.38 1.02 1.20 1.15 1.66 n.a. CPId 1.11 1.17 n.a. n.a. n.a. n.a. n.a. 1.24 1.33 1.17 1.05 n.a. CommentLithology RSWC: rotary sidewall core. DCM: dichloromethane. n.a.: not analysed. n.d.: not detected. KB: Kelly Bushing. GEUS Bulletin no 9 - 7 juli.pmd 07-07-2006, 14:1968 69 In order to preserve the samples intact for other inves- tigations, extracts were recovered from the surface and the immediately accessible pore spaces of the samples by im- mersing entire RSWCs into dichloromethane (DCM) for approximately 15 minutes in a glass beaker placed in an ultrasonification device. Rock fragments were separated by centrifugation, DCM was removed by evaporation, and the recovery was determined by weighing. Due to very low recoveries, no further preparative proce- dures were applied. Total extracts were re-dissolved in iso- octane and analysed by gas chromatography (GC) and by coupled gas chromatography – mass spectrometry (GC- MS). RSWC #1 (3543 m) was treated as described above, but a small chip containing a calcite-filled vug was re- moved from the sample. The vug was dissolved in dilute hydrochloric acid (HCl, 2N), and an organic extract was recovered by several stages of shaking the solute with dichloromethane (DCM) in a separatory funnel (Table 1). The basaltic counterpart of the vug was coarsely crushed and extracted for four hours in a Soxtec apparatus (1h reflux in DCM, followed by 3h rinsing). Asphaltenes were precipitated from both extracts by addition of 40-fold excess of n-pentane, and the maltenes were separated into saturated, aromatic and polar compounds by Medium Pressure Liquid Chromatography (MPLC), using a meth- od modified from Radke et al. (1980). Saturate fractions were analysed by GC and GC-MS. In order to remedy problems with a stuck pipe, diesel was added to the well at a depth of approximately 3100 m. Neither samples of the diesel, nor of the drilling mud, were preserved for analysis. Instead, drilling mud from bagged drill cutting samples collected before and after addition of diesel were analysed in order to check for pos- sible contamination. Approximately 10 g of sample (drill- ing mud plus cuttings) were ultrasonically extracted with DCM (app. 100 ml). Extracts were recovered by decanta- tion and centrifugation, and the DCM removed by evap- oration. Asphaltenes were precipitated by addition of 40- fold excess of n-pentane, with the maltenes being separat- ed into saturated, aromatic and polar components as de- scribed above. The saturate fractions were analysed by GC and GC-MS. An additional possible source of contamination was pipe dope, an anti-seize compound used when joining drilling rods. A sample of the pipe-dope used during drilling of the Lopra-1/1A re-entry well was supplied by DANOP and analysed using standard procedures for oil analysis. Asphaltenes were precipitated by addition of 40-fold ex- cess of n-pentane, and the maltenes were separated into saturated, aromatic and polar components as described above. Again the saturate fractions were analysed by GC and GC-MS. Gas chromatographic analyses were carried out by means of a Hewlett-Packard 5890 Series II plus gas chro- matograph, using splitless injection, a 25 HP-1 WCOT column and a flame ionisation detector (FID). Biological marker analyses were carried out by means of a Hewlett-Packard 5980A Series II gas chromatograph interfaced to a Hewlett-Packard 5971 mass selective de- tector (MSD) using splitless injection and a 25 m HP-5 WCOT column. RSWC #51, 2361 m RSWC #45, 2450 m RSWC #36, 2470 m RSWC #13, 3438 m RSWC #1, 3543 m RSWC #33, 2630 m RSWC #24, 3076 m, basalt diesel non-diesel Fig. 1. Gas chromatograms obtained from analysis of RSWCs rinsed in DCM. Approximate shapes and positions of ‘diesel’ and ‘non- diesel’ envelopes are shown in sample RSWC #51. Examples of ‘prominent unknowns’ are indicated by asterisks in sample RSWC #36, see Fig. 3. Note that all samples show some degree of diesel contamination, although this adulterant was not added until a depth of approximately 3100 m. This observation suggests that diesel contamination is pervasive throughout the entire uncased section of the well. GEUS Bulletin no 9 - 7 juli.pmd 07-07-2006, 14:1969 70 Results RSWCs rinsed in DCM Total extract recoveries were generally close to 1 mg, and no attempt was made to fractionate the residues. Gas chro- matograms of the bulk extracts (Fig. 1) show that most sam- ples contain a more or less well-developed series of n-al- kanes. The distributions are generally light-end skewed and unimodal, but two samples show evidence of bimo- dal distributions with higher proportions of waxy (22+) n-alkanes. Pristane/phytane ratios range from 1.34 to 1.61, and no odd- or even-number predominance is noted among the n-alkanes. In addition to n-alkanes, a number of prominent peaks of unknown identity are noted in all but one of the chromatograms, see below. Key parameters are listed in Table 1. Biological marker maturity indications are consistent, with homohopane and bishomohopane 22S/(22S + 22R) epimerisation ratios at equilibrium (i.e. close to 0.60), and C29 sterane 20S/(20S + 20R) epimerisation ratios slightly below equilibrium (i.e. slightly less than 0.52), indicating early to mid oil-window maturity (Table 2). Sterane and triterpane biological marker distributions display only minor variations among the samples. Representative ion fragmentograms m/z 191, m/z 217, and m/z 218 are shown in Fig. 2, with key biological marker parameter ratios tabu- lated in Tables 3, 4. Triterpane distributions generally com- prise fairly high proportions of tricyclic triterpanes. Among the pentacyclics the presence of 30-norhopanes plus 25- norhopanes is noted together with a peak eluting frac- tionally earlier than hopane. This peak is routinely assigned to 18α(H)-oleanane, a well-known marker of angiosperm higher plant inputs, and hence of source ages younger than the mid Cretaceous. However, scan-mode mass spectro- metric analysis could not confirm the identity of this com- pound. Rather the peak represents several co-eluting com- pounds, probably comprising oleanane, lupane, and one or more unknowns. Limited amounts of sample preclud- ed further investigation of this problem of identification. Norhopane to hopane ratios are close to unity and ex- tended hopanes are relatively abundant. Moreover, gam- macerane and notable proportions of hexahydrobenzoho- pane are present. Sterane distributions are very similar for all samples, comprising a clear predominance of C27 steranes over the C28 plus C29 steranes, but with the co-occurrence of C30 as well as of C26 steranes. Prominence of ββ-sterane epim- ers is noted in both the m/z 217 and m/z 218 ion frag- mentograms. A number of prominent unknowns are noted in the Pr is ta ne Ph yt an e n15 n20 n25 squalene m/z 218 sq 29 28 27 26 25 m/z 191 sq 1 2 3 4 5 6 7 8 9 10 11 12 15 16 19 20 21 22 23 13 14 1718 m/z 217 sq 25 27 29 28 S R 26 24 * * * * Fig. 2. Gas chromatogram and representative ion fragmentograms m/z 191, m/z 217 and m/z 218 (sample RSWC #45, 2450 m). Filled black peak labelled ‘sq’ is the contaminant squalene – a com- pound found on the skin of humans. Compound identification is shown in Table 5. GEUS Bulletin no 9 - 7 juli.pmd 07-07-2006, 14:1970 71 gas chromatograms as described above. Scan-mode mass spectrometric analyses of the extract obtained from RSWC #1, in which these compounds are particularly abundant, identifies: squalene and cholesterol; 2-ethylhexyl-phtha- late; vitamin A and various esters such as butyl-stearate, stearyl-palmitate and a series of similar compounds (Fig. 3). Calcite vug dissolved in HCl and basalt coarsely crushed Total extract recovery of the acid-digested calcite vug was only 1.1 mg, whereas solvent extraction of the tuff host- rock yielded 6.3 mg. The comparatively high extraction yield of the latter sample is probably due to the more effi- cient extraction procedure, i.e. crushing and Soxtec extrac- tion as opposed to ultrasonical extraction of entire samples. Gas chromatograms of saturate fractions of the two sub-samples (Fig. 4) are different. The extract recovered from the acid-digested calcite vug shows a strongly uni- modal n-alkane distribution, centred around C20, with notable light-end depletion and a rather poor signal-to- noise ratio. The basalt extract shows a well-developed un- imodal distribution of n-alkanes, centred around C17, and a clear odd-predominance in the C25–32 range. Pristane/phy- tane ratios are 1.02 and 1.20, respectively. Biological marker distributions are similar in the two samples, although the signal-to-noise ratio observed in the acid-digested sample is rather poor (Fig. 4). These distri- butions are significantly different from the picture pro- vided by other samples, including both sidewall cores (see above), drilling mud and pipe dope (see below). Triter- pane distributions comprise low proportions of tricyclic triterpanes. Amongst the pentacyclics the presence of 28,30-bisnorhopane is noted, while 30-norhopanes, 25- norhopanes, and ‘oleanane’ are absent, or cannot be iden- tified with any degree of certainty. H29/H30 ratios are close to 0.3 and extended hopanes are relatively scarce. Regular sterane distributions comprise a clear predomi- nance of C27 steranes over the C28 plus C29 homologues with the presence of C30 as well as C26 steranes. Drilling mud Total extract recovery from the drilling mud samples dif- fers widely before and after the addition of diesel to the drilling mud at a depth of approximately 3100 m bRKB (Table 1). This difference recurs in the gas chromatograms of saturate fractions of the two samples (Fig. 5). The sam- ple collected before addition of diesel yields a rather irreg- ular light-end skewed n-alkane distribution with high pro- portions of ‘Unresolved Complex Mixture’ (UCM). The sample collected after addition of diesel yields a well-de- fined, nearly symmetrical, unimodal n-alkane distribution, centred around C16. Pristane/phytane ratios are 1.15 and 1.66, respectively. Except for a minor enhancement of C27 diasteranes, and a slightly more pronounced enhancement of low molecular weight tricyclic triterpanes in the diesel-con- taining sample, biological marker distributions, however, are similar in the two samples and indistinguishable from Time (min.) 10 20 30 40 50 60 Abundance 0 2.4 x 107 2.0 x 107 Pr is ta ne Ph yt an e ? ? ? ? ? ? St ea ry l-p al m ita te C ho le st er ol Squalene Bu ty l-s te ar at e V ita m in -A ‘Diesel-envelope’ Common natural contaminants from ‘greasy fingers’ Various hand lotion and chap stick components Plasticiser for e.g. PVC plastic 1.6 x 107 1.2 x 107 8 x 106 4 x 106 2-Ethylhexyl-phthalate + + + + + + + + +++x x * * * Fig. 3. Full scan total ion fragmentogram, RSWC #1. The ap- proximate shape and position of a ‘diesel envelope’ are shown by shading. Interpretation of the origins of various contaminants is shown. S29 S/(S+R)a 0.49 0.48 0.46 0.43 0.48 0.49 0.47 0.44 0.51 0.45 0.50 0.42 Sample RSWC #51 RSWC #45 RSWC #36 RSWC #33 RSWC #24 RSWC #13 RSWC #1 RSWC #1 RSWC #1 Drilling mude Drilling mudf Pipe dope a C29 regular sterane αα20S/(αα20S+αα20R) epimer ratio. b C29 regular sterane ββ/(ββ+αα) epimer ratio. c homohopane 22S/(22S+22R) epimer ratio. d bishomohopane 22S/(22S+22R) epimer ratio. e before addition of diesel. f after addition of diesel. RSWC: rotary sidewall core. S29 βα/(ββ+αα)b 0.62 0.60 0.60 0.63 0.61 0.60 0.60 0.58 0.54 0.61 0.64 0.42 Table 2. Biological marker maturity data 0.61 0.59 0.60 0.59 0.56 0.57 0.59 0.58 0.58 0.59 0.59 0.65 H31 S/(S+R)c 0.63 0.61 0.60 0.63 0.59 0.60 0.59 0.59 0.61 0.62 0.59 0.57 H32 S/(S+R)d GEUS Bulletin no 9 - 7 juli.pmd 07-07-2006, 14:1971 72 the biological marker distributions yielded by RSWCs rinsed in DCM (Fig. 5). Pipe dope The gas chromatogram of the saturate fraction does not allow identification of any components, but simply shows a large hump of ‘Unresolved Complex Mixture’ (Fig. 6). Biological marker maturity parameters indicate early to mid-oil window maturity (Table 2). Triterpane distri- butions generally comprise high proportions of tricyclic triterpanes, and among the pentacyclics the presence of 30-norhopanes plus abundant 25-norhopanes is noted together with a peak eluting fractionally earlier than ho- pane, probably representing several co-eluting compounds. These possibly comprise oleanane, lupane and one or more unknowns (Fig. 6). The H29/H30 ratio is 1.12 with ex- tended hopanes being abundant. Gammacerane and no- table proportions of hexahydrobenzohopane are present. The distribution of regular steranes shows a clear pre- dominance of C27 steranes over the C28 and C29 homo- logues plus the presence of C30 as well as C26 steranes. Prominence of ββ-sterane epimers is noted in both the m/z 217 and m/z 218 ion fragmentograms. Discussion The amounts of extract recovered from the RSWC sam- ples are generally very low, which naturally limits the pos- sibilities for detailed studies and implementation of ex- tensive sample preparation techniques in order to opti- mise the quality of analytical data obtained. Furthermore, since total recoveries are low, even minor random con- tamination, which normally would be insignificant and ignored, may cause notable problems. Hence, considera- ble uncertainty is attached to the conclusions made on the basis of the analyses reported here. Analyses of mud samples collected before and after ad- dition of diesel show a profound influence of diesel on the content and distribution of n-alkanes, whereas the bio- logical marker characteristics, except for minor enhance- ment of C27 diasteranes and low molecular weight tricy- clic triterpanes, are largely unaffected. Hence for most practical purposes, this particular diesel distillate fraction will not severely influence any of the biomarker ratios. The biomarker distributions observed in the two mud samples are largely identical to the distributions yielded by DCM-rinsed RSWC samples. Based on gas chromatography data, most DCM-rinsed RSWC samples are seen to contain diesel, but some ex- tracts also contain longer chain-length n-alkanes and bio- logical markers which are unlikely to originate from die- sel contamination. In addition to diesel, contamination from various other sources is present in most samples: Sample RSWC #51 RSWC #45 RSWC #36 RSWC #33 RSWC #24 RSWC #13 RSWC #1 RSWC #1 RSWC #1 Drilling mudf Drilling mudg Pipe dope a C23 tricyclic terpane to hopane ratio. b Ts: 18α(H)-trisnorneohopane, Tm: trisnorhopane. c H28: 28,30-bisnorhopane, H29: norhopane. d H29: norhopane, H30: hopane. e OL: 18α(H)-oleanane, H30: hopane. f before addition of diesel. g after addition of diesel. n.a.: not analysed. RSWC: rotary sidewall core. Table 3. Key triterpane biological marker parameter ratio 0.00 0.14 0.18 0.14 0.52 0.80 0.36 0.00 0.00 0.03 0.04 0.16 OL/H30e 0.96 1.06 0.96 0.93 1.01 0.96 0.94 0.33 0.28 0.99 1.02 0.12 H29/H30d 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.35 0.37 0.00 0.00 0.00 H28/H29c 0.49 0.52 0.52 0.54 0.57 n.a. n.a. 0.43 0.44 0.51 0.51 0.41 Ts/(Ts+Tm)b 0.93 1.00 1.00 1.14 1.00 1.06 0.64 0.13 0.12 0.50 1.28 0.44 T23/H30a a (sum C27 diateranes)/(sum C27 regular steranes), m/z 217. b relative distribution of C27–29 regular steranes, based on αα20R epimers in m/z 217. c C27/C29 regular sterane ratio, based on αα20R epimers in m/z 217. d C30 regular steranes. e before addition of diesel. f after addition of diesel. RSWC: rotary sidewall core. Sample RSWC #51 RSWC #45 RSWC #36 RSWC #33 RSWC #24 RSWC #13 RSWC #1 RSWC #1 RSWC #1 Drilling mude Drilling mudf Pipe dope D27/S27a 1.33 1.27 1.39 1.53 1.20 1.22 0.95 1.04 1.06 0.81 1.19 0.47 Table 4. Key sterane biological marker parameter ratios 19.0 21.3 23.0 20.0 19.56 21.7 22.4 21.3 21.4 19.5 24.6 23.8 S28 (%)b 30.2 32.0 31.0 30.0 29.4 30.4 30.3 31.9 30.4 33.5 29.5 32.8 S29 (%)b 1.7 1.5 1.5 1.7 1.7 1.6 1.6 1.5 1.6 1.4 1.6 1.3 S27/S29c present present present present present present present present present present present present S30d 50.8 46.7 46.0 50.0 51.0 47.8 47.4 46.8 48.2 47.0 45.9 43.3 S27 (%)b GEUS Bulletin no 9 - 7 juli.pmd 07-07-2006, 14:1972 73 pr is ta ne ph yt an e n15 n20 n25 pr is ta ne ph yt an e n15 n20 n25 m/z 191 28 ,3 0- bi sn or ho pa ne 3 4 9 10 11 15 16 17 19 20 21 22 13 m/z 191 28 ,3 0- bi sn or ho pa ne 3 4 9 10 11 15 16 19 20 21 22 13 m/z 218 28 2726 25 m/z 217 24 25 27 28 S R 26 * * * * m/z 218 28 29 2726 25 m/z 217 24 25 27 28 S R 26 * * * * Fig. 4. Gas chromatograms and ion fragmentograms m/z 191, m/z 217 and m/z 218, RSWC #1, calcite vug dissolved in HCl (left), and its coarsely crushed tuff host rock (right). Note mutual similarity of biological marker distributions and differences when compared to distributions shown in Figs 2, 5, 6. Compound identification is shown in Table 5. GEUS Bulletin no 9 - 7 juli.pmd 07-07-2006, 14:1973 74 m/z 218 m/z 218 29 2827 26 25 29 28 27 26 25 m/z 191 1 2 3 4 5 6 7 8 9 10 11 12 15 16 19 20 21 22 23 13 14 17 18 m/z 217 25 27 28 29 S R 26 24 * * * * m/z 191 1 2 3 4 5 6 7 8 9 10 11 12 15 16 19 20 21 22 23 13 14 17 18 m/z 217 25 27 28 29 S R 26 24 * * * * UCMpr is ta ne ph yt an e n15 n20 n25 pr is ta ne ph yt an e n15 n20 n25 Fig. 5. Gas chromatograms and ion fragmentograms m/z 191, m/z 217 and m/z 218, extract of drilling mud before addition of diesel (left), and extract of drilling mud after addition of diesel (right). UCM, Unresolved Complex Mixture. Note profound influence of diesel on the n-alkane distribution, and the lack of, or limited effect on, biological marker distributions. Compound identification is shown in Table 5. GEUS Bulletin no 9 - 7 juli.pmd 07-07-2006, 14:1974 75 squalene and cholesterol are commonly occurring natural compounds present on the skin of humans, for instance on the hands (‘naturally greasy fingers’); 2-ethylhexyl-phtha- late is a widely used plasticiser for various polymers; vita- min A, esters such as butyl-stearate, stearyl-palmitate and similar compounds detected in the sample are commonly used in cosmetics, including hand lotion and chap-stick. The absence of both acyclic isoprenoids and n-alkanes from pipe dope plus the presence of abundant 25-norho- panes suggest that this product is based on a heavily bio- degraded oil (corresponding to level 6 of Peters & Moldow- an (1993)). Based on the presence of 30-norhopanes, which is also manifest in norhopane to hopane ratio close to unity plus the prominence of ββ-sterane epimers and comparatively high proportions of homohopanes (in par- ticular tetrakishomohopane to pentakishomohopane ratio close to unity), this oil was presumably generated from a marine marly source rock, deposited in a highly anoxic environment. In addition, the presence of angiosperm higher land plant markers such as 18α(H)-oleanane sug- gest a source age younger than the mid-Cretaceous. Nor- mal and acyclic isoprenoid alkanes are absent, but the bio- logical marker distribution shows clear similarities to the distributions yielded by DCM-rinsed RSWCs discussed above, such that the presence of contamination from pipe dope, in addition to adulteration from other sources, seems obvious. However, differences are observed: pipe dope contains very high proportions of 25-norhopanes, where- as the proportion in the RSWCs are but minor; the rela- tive abundance of Ts and Tm is reversed in pipe dope com- pared to DCM-rinsed RSWC samples. Similarly, the pro- portion of C27 diasteranes relative to C27 regular steranes is much lower in pipe dope extracts than in the DCM- rinsed RSWC samples. The latter feature may, however, be wholly or partly caused by the addition of diesel, which was shown above to result in minor enhancement of low boiling-range tricyclics and diasteranes relative to non- contaminated samples. In summary, DCM-rinsed samples are contaminated by a variety of compounds originating from several sourc- es, including diesel, pipe dope, plasticisers, naturally greasy fingers and cosmetics, possibly hand lotion and/or chap stick. However, the samples also contain petroleum com- ponents that do not seem to originate from these sources of contamination. An origin from other sources of con- tamination or from indigenous crude oil is conceivable. Laier et al. (1997) show the presence of traces of heavy pe- troleum hydrocarbons and wax in samples from the origi- nal Lopra-1 well, prior to deepening. The ‘unexplainable’ petroleum components found in DCM-rinsed samples from the deepened well may represent similar occurrences. Level n15 a b n20 n25 sq n30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 C15 normal alkane pristane phytane C20 normal alkane C25 normal alkane squalene C30 normal alkane C28 tricyclic terpanes (2 isomers) C29 tricyclic terpanes (2 isomers) Ts = trisnorneohopane Tm = trisnorhopane C30 tricyclic terpanes (2 isomers) C28 25,30 bisnorhopane, coeluting with "5" C31 tricyclic terpanes (2 isomers) C29 25-norhopane, partially coeluting with "7" norhopane = 30-norhopane 29Ts = norneohopane normoretane mixture of oleanane, lupane and unknown hopane C30 30-norhopane moretane homohopane, 22S and 22R isomers gammacerane C31 hexahydrobenzohopane homomoretane bishomohopane, 22S and 22R isomers trishomohopane, 22S and 22R isomers tetrakishomohopane, 22S and 22R isomers pentakishomohopane, 22S and 22R isomers C27 diasteranes, 4 isomers labelled with asterisks C26 regular steranes, ββ isomers only C27 regular steranes, αα and ββ isomers C28 regular steranes, ββ isomers only C29 regular steranes, ααS, ααR and ββ isomers C30 regular steranes, ββ isomers only Table 5. Compound identification key Compound Biological marker, in particular triterpane, distributions in extracts recovered from DCM-rinsing of RSWC sam- ples from the deepened Lopra-1 well show a number of striking similarities to distributions yielded by certain sam- ples collected in the original Lopra-1 well prior to re-entry. These samples include petroleum extracted from water samples in 1992 (Laier et al. 1997), and a sample collect- ed from slurry sticking to a VSP-tool, which was lowered into the hole in 1994. The biological marker characteris- tics include features such as the presence of 30-norhopanes, 25-norhopanes, gammacerane, hexahydrobenzohopane, H29/H30 ratios close to or greater than 1 and a high rel- ative abundance of extended hopanes plus prominence of the ββ-sterane epimers. Minor differences between pipe dope and the ‘slurry’ and the oil film are also observed, principally with respect to the presence of n-alkanes and the proportions of C28 regular steranes relative to the pro- GEUS Bulletin no 9 - 7 juli.pmd 07-07-2006, 14:1975 76 gest the presence of pipe dope contamination in the orig- inal borehole as well, the differences being accounted for by presumed differences between the pipe dopes used in the original and the deepened Lopra wells. Furthermore, the presence of a mid-Cretaceous or younger marly, anox- ic marine source rock in the area, as implied by the geo- chemical data, seems geologically problematical. Carbon- ate or marly source rocks typically occur in lower latitude regions, i.e. areas within and close to the arid tropical belts (Tissot & Welte 1984). It is estimated that during the Cretaceous, the Faroe Islands area was situated at 40–45°N, and northward movement has prevailed since then (Ha- bicht 1979; Scotese et al. 1988). The biological marker characteristics of the extracts recovered from acid-digestion of a calcite-filled vug, and from its crushed tuff host-rock are totally dissimilar to the characteristics shown by all other samples from the well, including pipe dope. Principally, the distribution of triterpanes serves to distinguish these two samples from the remainder of the samples analysed. Hence, the fol- lowing characteristic are noteworthy: lower proportions of tricyclic triterpanes, low norhopane to hopane ratio and the presence of 28,30-bisnorhopane. The signal-to-noise ratio is comparatively poor, but the overall biological marker characteristics show some similarities to oils gen- erated from Upper Jurassic Kimmeridge Clay Formation sediments and their equivalents. This source system is known to be present in West of Shetlands basins (e.g. Scotchman et al. 1998) as well as in the North Sea basins, and can be prognosed for Faroese waters. Glassley (2006, this volume) estimates that the maxi- mum temperature reached at TD of the Lopra-1/1A re- entry well was 200°C. Provided that this estimate is cor- rect, the temperature may be too high to allow preserva- tion of higher molecular weight liquid hydrocarbons if maintained over prolonged periods of time. The maxi- mum temperature actually recorded in the well was 98°C, and assuming that the hydrocarbons found in RSWC #1 represent a thermogenic natural petroleum, this may have entered the tuff at a temperature lower than the maxi- mum estimated by Glassley (2006, this volume), proba- bly in connection with the formation of the calcite vug. Petroleum-bearing fluid inclusions have been observed in the same sample (Konnerup-Madsen 2006, this volume). Based on fluorescence colours (orange-yellow to green), an API gravity of 20–35 is estimated. No homogenisation temperature data have been recorded for the petroleum- bearing inclusions, but data from non-hydrocarbon bear- ing inclusions fall in the range 101–186°C, with the higher temperatures probably being caused by partial decrepita- tion. Hence, entrapment temperatures were probably con- 1 2 3 4 5 6 7 8 9 10 11 12 15 16 19 20 21 22 23 25 27 29 29 28 27 26 25 28 S R 26 13 14 17 18 UCM 24 * * * * m/z 191 m/z 217 m/z 218 Fig. 6. Gas chromatogram and ion fragmentograms m/z 191, m/z 217 and m/z 218, pipe dope. UCM, Unresolved Complex Mix- ture. Compound identification is shown in Table 5. portions of C27 and C29 regular steranes. However, the striking similarities and the presence of somewhat unusual components in pipe dope, as well as in the samples collect- ed from the Lopra-1 well, prior to deepening, may sug- GEUS Bulletin no 9 - 7 juli.pmd 07-07-2006, 14:1976 77 siderably lower than the maximum temperature as esti- mated by Glassley (2006, this volume), and it is conceiv- able that the signal obtained represents indigenous crude oil. If so, this observation is very encouraging for future exploration in Faroese territorial areas. Conclusions All samples are to variable degrees contaminated by a number of adulterants of different origins. The adulter- ants include: 1. n-alkanes and other petroleum components originat- ing from commercial diesel fuel; 2. pipe dope, an anti-seize compound used, for instance, when joining drilling rods; 3. squalene and cholesterol originating from naturally greasy fingers; 4. vitamin A and various esters used in cosmetics, includ- ing chap stick and hand lotion; 5. 2-ethylhexyl-phthalate, a widely used plasticiser for polymers/plastics. A number of striking similarities in biological marker dis- tributions between pipe dope and samples collected in the well prior to re-entry and deepening may suggest the pres- ence of pipe dope contamination in the original well too, in addition to the presence of traces of petroleum hydro- carbons as shown by Laier et al. (1997). Organic extracts recovered from dissolution of a cal- cite vug in RSWC #1 (3543 m) and from its tuff host rock yield biological marker distributions different from all other samples collected in the Lopra-1 well. The distri- bution hints at generation from a marine anoxic shale source rock similar to the Kimmeridge Clay Formation and equivalents known from surrounding areas. It is con- ceivable that this organic extract represents an indigenous thermogenic petroleum, in particular since oil-bearing fluid inclusions have been observed in the same sample. Acknowledgements DANOP kindly supplied a sample of the pipe dope used during drilling of the well. Ditte Kiel-Dühring assisted in the preparation and analysis of the samples. Troels Laier and the reviewers Dr. R. Burwood and Dr. G. van Graas provided helpful comments and suggestions, which sig- nificantly improved the manuscript. References Boldreel, L.O. 2006: Wire-line log-based stratigraphy of flood ba- salts from the Lopra-1/1A well, Faroe Islands. Geological Survey of Denmark and Greenland Bulletin 9, 7–22 (this volume). Glassley, W.E. 2006: Mineralogical and thermodynamic constraints on Palaeogene palaeotemperature conditions during low-grade metamorphism of basaltic lavas recovered from the Lopra-1/1A deep hole, Faroe Islands. Geological Survey of Denmark and Greenland Bulletin 9, 109–118 (this volume). Habicht, J.K.A. 1979: Paleoclimate, paleomagnetism, and conti- nental drift. AAPG Studies in Geology 9, 31 pp. + maps. Heinesen, M.V., Larsen, A.R. & Sørensen, K. 2006: Introduction. Geological Survey of Denmark and Greenland Bulletin 9, 5–6 (this volume). Jacobsen, O.S. & Laier, T. 1984: Analysis of gas and water samples from the Vestmanna-1 and Lopra-1 wells, Faeroe Islands. In: Berthelsen, O., Noe-Nygaard, A. & Rasmussen, J. (eds): The deep drilling project 1980–91 in the Faeroe Islands. Annales Societatis Scientiarum Færoensis Supplementum IX, 149–155. Tórshavn: Føroya Frodskaparfelag. Jørgensen, O. 1984: Zeolite zones in the basaltic lavas of the Faeroe Islands. In: Berthelsen, O., Noe-Nygaard, A. & Rasmussen, J. (eds): The deep drilling project 1980–91 in the Faeroe Islands. Annales Societatis Scientiarum Færoensis Supplementum IX, 71– 91. Tórshavn: Føroya Frodskaparfelag. Konnerup-Madsen, J. 2006: A reconnaissance study of fluid inclu- sions in fracture filling quartz and calcite from the Lopra-1/1A well, Faroe Islands. Geological Survey of Denmark and Green- land Bulletin 9, 119–122 (this volume). Laier, T., Nytoft, H.P., Jørgensen, O. & Isaksen, G.H. 1997: Hy- drocarbon traces in the Tertiary basalts of the Faeroe Islands. Marine and Petroleum Geology 14, 257–266. Peters, K.E. & Moldowan, J.M. 1993: The biomarker guide, 363 pp. New Jersey: Prentice Hall. Radke, M., Willsch, H. & Welte, D.H. 1980: Preparative hydro- carbon group type determination by automated Medium Pres- sure Liquid Chromatography. Analytical Chemistry 52, 406–411. Scotchman, I., Griffith, C.E., Holmes, A.J. & Jones, D.M. 1998: The Jurassic petroleum system north and west of Britain: a geo- chemical oil–source correlation study. Organic Geochemistry 29, 671–700. Scotese, C.R., Gahagan, L.M. & Larson, R.L. 1988: Plate tectonic reconstruction of the Cretaceous and Cenozoic ocean basins. Tectonophysics 155, 27–48. Tissot, B. & Welte, D.H. 1984: Petroleum formation and occur- rence, 2nd edition, 699 pp. Berlin: Springer Verlag. Waagstein, R. 1988: Structure, composition and age of the Faeroe basalt plateau. In: Morton, A.C. & Parson, L.M. (eds): Early Tertiary volcanism and the opening of the NE Atlantic. Geolog- ical Society Special Publication (London) 39, 225–238. Waagstein, R., Hald, N., Jørgensen, O., Nielsen, P.H., Noe-Ny- gaard, A., Rasmussen, J. & Schönharting, G. 1984: Deep drill- ing on the Faeroe Islands. Bulletin of the Geological Society of Denmark 32, 133–138. Manuscipt received 16 June 2000; revision accepted 9 November 2000. GEUS Bulletin no 9 - 7 juli.pmd 07-07-2006, 14:1977