Determination of ruthenium in photographic emulsions – development and comparison of different sample treatments and mass spectrometric methods

Different sample treatment procedures were combined with inductively coupled plasma mass spectrometry (ICP-MS) and negative thermal ionisation mass spectrometry (NTI-MS) for the determination of ruthenium traces in photographic emulsions. Dissolution of the samples in concentrated ammonia solution was used in connection with ICP-MS by external calibration, which has the advantage of a simple sample preparation technique but introduces high amounts of the silver matrix into the mass spectrometer. On the other hand, isotope dilution mass spectrometry (IDMS) with an enriched ⁹⁹Ru spike solution was applied for ICP-MS and NTI-MS measurements, respectively, in connection with a significant reduction of the matrix by AgCl precipitation. In these cases loss of ruthenium by the AgCl precipitate has no effect on the analytical result. The results of the different methods agreed usually well analysing ruthenium traces in the range of 0.1–10 μg per gram emulsion. The detection limits obtained were 4 ng/g for ICP-IDMS, 20 ng/g for NTI-IDMS, and 15 ng/g for ICP-MS with external calibration. Differences in the results between the different methods could mainly be attributed to sample inhomogeneities. ICP-IDMS with silver matrix reduction by AgCl precipitation is recommended as a routine method, NTI-IDMS with the corresponding sample treatment as a calibration method.     

Evaluation of the combined measurement uncertainty in isotope dilution by MC-ICP-MS

The combination of metrological weighing, the measurement of isotope amount ratios by a multicollector inductively coupled plasma mass spectrometer (MC-ICP-MS) and the use of high-purity reference materials are the cornerstones to achieve improved results for the amount content of lead in wine by the reversed isotope dilution technique. Isotope dilution mass spectrometry (IDMS) and reversed IDMS have the potential to be a so-called primary method, with which close comparability and well-stated combined measurement uncertainties can be obtained.This work describes the detailed uncertainty budget determination using the ISO-GUM approach. The traces of lead in wine were separated from the matrix by ion exchange chromatography after HNO(3)/H(2)O(2) microwave digestion. The thallium isotope amount ratio ( n((205)Tl)/ n((203)Tl)) was used to correct for mass discrimination using an exponential model approach. The corrected lead isotope amount ratio n((206)Pb)/ n((208)Pb) for the isotopic standard SRM 981 measured in our laboratory was compared with ratio values considered to be the least uncertain. The result has been compared in a so-called pilot study "lead in wine" organised by the CCQM (Comité Consultatif pour la Quantité de Matière, BIPM, Paris; the highest measurement authority for analytical chemical measurements).The result for the lead amount content k(Pb) and the corresponding expanded uncertainty U given by our laboratory was:k(Pb)=1.329 x 10-10mol g-1 (amount content of lead in wine)U[k(Pb)]=1.0 x 10-12mol g-1 (expanded uncertainty U=kxuc, k=2)The uncertainty of the main influence parameter of the combined measurement uncertainty was determined to be the isotope amount ratio R(206,B) of the blend between the enriched spike and the sample.     

A comparison of double-focusing sector field ICP-MS, ICP-OES and octopole collision cell ICP-MS for the high-accuracy determination of calcium in human serum

Human serum is routinely measured for total calcium content in clinical studies. A definitive high-accuracy and low-uncertainty method is required for reference measurements to underpin medical diagnoses. This study presents a novel octopole collision cell ICP-MS, high-accuracy, methodology and comparison of that technique with double-focusing sector field ICP-MS and an ICP-OES method. Double-matched isotope dilution mass spectrometry (IDMS) was employed for ICP-MS techniques and an exact matching bracketing technique using scandium as an internal standard was used for ICP-OES analysis. Medium resolution mode was utilised for double-focusing sector field ICP-MS analysis to resolve the dominant interferences on the ⁴⁴Ca/⁴²Ca isotope pair. Hydrogen reaction gas was employed to chemically resolve a number of polyatomic interferences predominantly through charge transfer reactions in the octopole collision cell. Comparison data presented for NIST CRM 909b human serum analysis from all three techniques demonstrates highest accuracy (99.6%) and lowest uncertainty (1.1%) for octopole collision cell ICP-MS. Data from ICP-OES using a non-IDMS technique produces comparably accurate data and low-uncertainties. The much higher total expanded uncertainties for double-focusing sector field ICP-MS compared with octopole collision cell data are explained by lower precision on the measurement of the ⁴⁴Ca/⁴²Ca isotope ratio. Data for octopole collision cell ICP-MS submitted for an international blind trial comparison (CCQM K-14) demonstrated excellent agreement with the mean of all participants with a low expanded uncertainty.     

The optimal amount of isotopic spike solution for ultratrace analysis by isotope dilution mass spectrometry

 Inductively coupled plasma mass spectrometry (ICP-MS) and high resolution inductively coupled plasma mass spectrometry (HR-ICP-MS) are powerful methods of determining metallic impurities in the low- and sub-ppt level in process media such as ultra-pure water used in semiconductor and wafer manufacturing. By using mass spectrometers for analysis, an isotope dilution analysis (IDMS) is possible. The reproducibility of an IDMS is unmatched. For concentration levels near the instrument detection limit a novel method is reported to find the optimal amount of isotopic spike solution. This optimal value can be derived by the law of propagation of uncertainty combined with the Poisson statistics of the measured number of counts. Generally, an excess of isotopic spike solution should be used to provide results of lowest possible uncertainty. The results are presented in a diagram for easy practical use. Received: 14 October 1997 · Accepted: 13 February 1998     

Accurate determination of element species by on-line coupling of chromatographic systems with ICP-MS using isotope dilution technique

The instrumental design for coupling different liquid chromatographic systems such as ion, reversed phase, and size exclusion chromatography as well as capillary gas chromatography, with ICP-MS for the determination of element species is described. For accurate analyses obtaining ‘real time’ concentrations of chromatographic peaks, the isotope dilution mass spectrometric (IDMS) technique is applied. Two different spiking modes are possible, one using species-specific and another one using species-unspecific spike solutions of isotope-enriched labelled compounds. The species-specific mode is only possible for element species well defined in their structure and composition, for example iodate or selenite, whereas the species-unspecific mode must be applied in all cases where the structure and composition of the species is unknown, for example, for metal complexes with humic substances. For accurate determinations by the isotope dilution technique the mass discrimination effect must also be taken into account. Iodate, iodide and organoiodine species, including those of humic substances, have been analysed in mineral, drinking and environmental water samples by coupling different liquid chromatographic methods with ICP-IDMS. Heavy metal complexes with humic substances in water samples of different origin have been characterized by size exclusion/ICP-IDMS. The possibilities of determining different environmental selenium species are discussed and the results for the analysis of selenite and selenate, which has been carried out by GC/ICP-IDMS after converting these species into a volatile piazselenol compound, are presented.     

Impact of pump flow fluctuations on post column online ID-ICP-MS

In post column online isotope dilution mass spectrometry (IDMS), the stability of the spike mass flow is a key element. Changes in viscosity or fluctuations in the pump rate of the peristaltic pump may affect the results of post column online IDMS measurements. It was shown by simulating random fluctuations and studying the changes in the resulting integrals of the isotope ratio chromatogram of the sample that even small fluctuations, observable when using peristaltic pumps, can influence the result and especially its uncertainty. The use of a balance to continuously monitor the mass flow of the spike during the measurement which we presented in a previous publication allows now to correct the isotope ratio chromatogram for these fluctuations. Subsequently, the simulated effect was verified experimentally for the determination of Se-Met in the human serum reference material BCR 637, where the corrected mass fraction was plainly closer to the mass fraction obtained by species specific IDMS. Additional attention was paid to the fact that there is a time shift between the observation of the fluctuations in the pump rate and the detection of these fluctuations in the ICP-MS.     

Application of the combination of isotope ratio monitoring with isotope dilution mass spectrometry to the determination of glucose in serum

Isotope ratio monitoring combined with n((13)C)/n((12)C) isotope dilution mass spectrometry (IRM/IDMS) provides results of low uncertainty of the order of 0.1% if it is applied to the analysis of simple mixtures as found in organic chemistry, even if only low (13)C spike additives to the sample are used. If the method is applied to the analysis of systems that require large-scale sample preparation prior to the measurement, such as the determination of glucose in serum, the results obtained exhibit a higher uncertainty that is comparable to that of the conventional gas chromatography/isotope dilution mass spectrometry (GC/IDMS) method. The reason for this observation is that the small contribution that the IRM/IDMS method makes to the uncertainty budget of the result is superimposed on a large contribution due to the sample preparation. It appears therefore that the IRM/IDMS method has no advantage over the conventional GC/IDMS method. However, if a series of measurements is carried out, and if a suitable experimental design is chosen, the IRM/IDMS method can provide valuable additional information. The influence of sample preparation on each individual result can be quantified as its deviation from the average value of all results of the series. From these data conclusions can be drawn for an improvement in sample preparation.     

Low uncertainty reverse isotope dilution ICP-MS applied to certifying an isotopically enriched Cd candidate reference material: A case study

An analytical method is presented based on reverse isotope dilution single detector inductively coupled plasma magnetic sector mass spectrometry (ID-ICP-SMS) and applied to the specific case of the certification of a (111)Cd enriched candidate Cd spike calibration material (nominal mass fraction 10 mg kg(-1) in 5% HNO3 solution). Uncertainty propagation was used as a tool for both determining the analytical approach and validating it. The robustness of close to "exact matching" reverse IDMS to correction of measured isotope intensities for multiplicative (mass discrimination) and (semi)additive effects (dead time, instrumental background, and isobaric interference) is discussed. The very low experimental relative standard deviation of the mean (0.08%) of eight replicate determinations indicated that all significant sources of uncertainty had probably been taken into account for the estimation of the final combined uncertainty statement (U(c) = 0.17%, k = 1). IRMM-621 was used as comparator. Uncertainties on IUPAC isotopic abundances of 111Cd and 112Cd, for the natural Cd solution involved between the two enriched materials, formed nearly 60% of U(c). The repeatability of the isotope ratio measurements contributed less than 10%. Correction for procedural blank necessitated somewhat unusual calculations (potential contamination of an enriched material with natural Cd). The procedure also involved a quadrupole based ICP-MS judged to be appropriate for the characterization of the isotopic composition. For comparison purposes, direct IDMS results are simulated using identical experimental input data. Finally, a significant background signal in the 106-116 mass region, observed only with the magnetic sector instrument, was attributed to argon based isobaric interferences.     

Elemental labelling combined with liquid chromatography inductively coupled plasma mass spectrometry for quantification of biomolecules: A review

This article reviews novel quantification concepts where elemental labelling is combined with flow injection inductively coupled plasma mass spectrometry (FI-ICP-MS) or liquid chromatography inductively coupled plasma mass spectrometry (LC–ICP-MS), and employed for quantification of biomolecules such as proteins, peptides and related molecules in challenging sample matrices. In the first sections an overview on general aspects of biomolecule quantification, as well as of labelling will be presented emphasizing the potential, which lies in such methodological approaches. In this context, ICP-MS as detector provides high sensitivity, selectivity and robustness in biological samples and offers the capability for multiplexing and isotope dilution mass spectrometry (IDMS). Fundamental methodology of elemental labelling will be highlighted and analytical, as well as biomedical applications will be presented. A special focus will lie on established applications underlining benefits and bottlenecks of such approaches for the implementation in real life analysis. Key research made in this field will be summarized and a perspective for future developments including sophisticated and innovative applications will given.     

New mathematical models with associated equations for isotope dilution mass spectrometry (IDMS)

Vector models which progressively lead to a general model for isotope dilution mass spectrometry (IDMS) are presented for the case of two 'monitor isotopes' and one blend involved. They enable one to find the boundary conditions for performing IDMS, and cover the cases of highly enriched isotopes, radioactive isotopes and ratios that are given with different denominator. The models identify the key measurements in their simplest form as well as the conditions which minimise the measurement effort and in some cases the propagated measurement uncertainties. The equations are discussed and compared with other published IDMS equations. Combined with discussion on fundamental aspects of IDMS, this results in an even more 'general' but also more complex IDMS equation.     

Contribution of ICP-IDMS to the certification of antimony implanted in a silicon wafer - comparison with RBS and INAA results

A thin-layer reference material for surface and near-surface analytical methods was produced and certified. The surface density of the implanted Sb layer was determined by Rutherford backscattering spectrometry (RBS), instrumental neutron activation analysis (INAA), and inductively coupled plasma isotope dilution mass spectrometry (ICP-IDMS) equipped with a multi-collector. The isotopic abundances of Sb (121Sb and 123Sb) were determined by multi-collector ICP-MS and INAA. ICP-IDMS measurements are discussed in detail in this paper. All methods produced values traceable to the SI and are accompanied by a complete uncertainty budget. The homogeneity of the material was measured with RBS. From these measurements the standard uncertainty due to possible inhomogeneities was estimated to be less than 0.78% for fractions of the area increments down to 0.75 mm2 in size. Excellent agreement between the results of the three different methods was found. For the surface density of implanted Sb atoms the unweighted mean value of the means of four data sets is 4.81 x 10(16) cm(-2) with an expanded uncertainty (coverage factor k = 2) of 0.09 x 10(16) cm(-2). For the isotope amount ratio R (121Sb/123Sb) the unweighted mean value of the means of two data sets is 1.435 with an expanded uncertainty (coverage factor k = 2) of 0.006.     

Improving species-specific IDMS: the advantages of triple IDMS

Triple isotope dilution mass spectrometry (triple IDMS) has been applied for the first time on protein quantification, especially on transferrin. Transferrin as an acute phase protein is a marker for several inflammation processes in the human body. Therefore, in Germany, the accurate and precise measurement of this important analyte is required. In this work, a new approach to triple IDMS is described and compared to double IDMS. Also, complete uncertainty budgets for both methods were set up to demonstrate the ability of this method to be used as a reference procedure. The relative expanded uncertainty (k?=?2) for triple IDMS (3.6 %) is smaller than the one for double IDMS (4.0 %). The content of transferrin found in the human serum reference material ERM-DA470k/IFCC ((2.41?±?0.08) g/kg) with both methods was in good agreement with each other and with the certificate. For triple IDMS ((2.426?±?0.086) g/kg) and for double IDMS ((2.317?±?0.092) g/kg), transferrin was determined. Although triple IDMS is a little more time consuming compared to double IDMS, there is the advantage that the isotopic composition of the spike material does not have to be determined. This is very useful especially in case of a marginal isotopic enrichment in the spike or problems with the accurate measurement of the spike isotope ratio.

微量气体大小球三步同位素稀释质谱法理论探讨与实验模型设计

提出了一种适用于微量气体定量分析的大小球三步同位素稀释质谱法,并给出其实验模型和不确定度理论分析结果。通过引入三步稀释过程,不涉及稀释剂的纯度、丰度、添加质量等方面的数据,仅需测定各步稀释剂质量之比,在其中的两步反稀释过程中使用了基标准纯物质,从而使微量气体的定量分析结果可溯源至基标准纯物质。该方法解决了已有同位素稀释质谱法在微量气体定量分析中的难题。     

Comparison of heavy metal analyses in hydrofluoric acid used in microelectronic industry by ICP-MS and thermal ionization isotope dilution mass spectrometry

An isotope dilution mass spectrometric (IDMS) method with the thermal ionization (TI) technique has been developed for the determination of trace impurities of Cr, Fe, Ni, Cu, Zn, Ag, Cd, Tl, Pb, Th, and U in high-purity HF (50% by weight) used in the semiconductor industry. The evaporation step of the HF solution was carried out in an apparatus which did not significantly contribute to contaminations of the heavy metals to be analysed. This apparatus allowed fast evaporation of the HF solution of up to 200 ml/h and therefore also a fast trace heavy metal/matrix separation was carried out. The evaporation step was also used in connection with inductively coupled plasma mass spectrometry (ICP-MS) when applying the isotope dilution technique and an external calibration for quantification, respectively. The detection limits for TI-IDMS were (in pg/g): Cr=30, Fe=400, Ni=70, Cu=20, Zn=1100, Ag=70, Cd=10, Tl=1, Pb=16, Th=3, and U=1. With ICP-MS in combination with the evaporation step, detection limits of less than 50 pg/g have been achieved for Cr, Ni, and Zn and of <5 pg/g for the other elements except Fe, which could not be determined in concentrations less than 100 ng/g. On the other hand, the detection limits were much higher when the HF matrix was not removed before measuring by ICP-MS. A comparison of the different ICP-MS methods (isotope dilution technique and external calibration for both HF evaporated samples and those with HF matrix) with the results of TI-IDMS has been carried out. An excellent agreement was achieved between the results of TI-IDMS and the two ICP-MS methods using the HF evaporation step, whereas the ICP-MS techniques without HF evaporation essentially deviated from these results. Fe was the only trace element of all investigated heavy metals which could only be analysed by TI-IDMS in high purity HF in a concentration of about 3 ng/g. Although ICP-MS with isotope dilution and external calibration resulted in comparable analytical data, the ICP-IDMS method has some practical advantages such as time-saving and more reliable results.     

Direct μ-flow injection isotope dilution ICP-MS for the determination of heavy metals in oil samples

The determination of trace elements in oil samples and their products is of high interest as their presence significantly affects refinery processes and the environment by possible impact of their combustion products. In this context, inductively coupled plasma mass spectrometry (ICP-MS) plays an important role due to its outstanding analytical properties in the quantification of trace elements. In this work, we present the accurate and precise determination of selected heavy metals in oil samples by making use of the combination of μ-flow direct injection and isotope dilution ICP-MS (ICP-IDMS). Spike solutions of ⁶²Ni, ⁹⁷Mo, ¹¹⁷Sn and ²⁰⁶Pb were prepared in an organic solvent, mixed directly with the diluted oil samples and tested to be fit for purpose for the intended ID approach. The analysis of real samples revealed strong matrix effects affecting the ICP-MS sensitivity, but not the isotope ratio measurements, so that accurate results are obtained by ICP-IDMS. Typical relative standard deviations were about 15% for peak area and peak height measurements, whereas the isotope ratios were not significantly affected (RSD < 2%). The developed method was validated by the analysis of a metallo-organic multi-element standard (SCP-21, typically applied as a calibration standard) and the standard reference material SRM1084a (wear metals in lubricating oil). The obtained results were in excellent agreement with the certified values (recoveries between 98% and 102%), so the proposed methodology of combining μ-flow direct injection and ICP-IDMS can be regarded as a new tool for the matrix-independent, multi-element and reliable determination of trace elements in oil and related organic liquids.     

Achieving traceable chemical measurements: inter-laboratory evaluation of a simplified technique for isotope dilution mass spectrometry. Part 2. Methodology for high accuracy analysis of organic analytes

A high accuracy measurement procedure developed and validated at LGC has been transferred to a number of expert UK laboratories, and their experience in applying the technique has been evaluated by inter-laboratory comparisons. It is an “exact matching” calibration procedure for analysis of organic analytes using isotope dilution mass spectrometry (IDMS). This calibration procedure uses a calibration blend and a sample blend with closely matched isotope amount ratios, and is an iterative process, culminating in the calibration blend and sample blend having identical isotope amount ratios. It is capable of high accuracy, since systematic errors in the determination of the isotope amount ratios are cancelled out. A series of four inter-laboratory comparisons of increasing difficulty were carried out involving a number of expert laboratories. The first three comparisons used gas chromatography mass spectrometry (GC–MS) analysis of the pesticide metabolite (pp′-dichlorodiphenyl) dichloroethylene (pp′-DDE), involving both conventional calibration and IDMS exact matching procedures for pp′-DDE in a solvent and a complex liquid matrix (corn oil). The fourth comparisons utilised liquid chromatography mass spectrometry (LC–MS) and involved the analysis of sulphamethazine (4-amino-N-(4,6 dimethyl-2 pyrimidinyl) benzenesulphonamide) in solvent using IDMS and conventional calibration techniques. Following the first trial, a workshop for participants was held on the use of the exact matching procedure together with a short course on uncertainty estimation. The results of the comparisons clearly showed the superior accuracy of using IDMS with the exact matching procedure for both GC–MS and LC–MS applications. These comparisons and the workshop have enabled the methodology to be transferred to UK industry, helping to improve UK measurement capability.

     

Achieving traceable chemical measurements: inter-laboratory evaluation of a simplified technique for isotope dilution mass spectrometry. Part 2. Methodology for high accuracy analysis of organic analytes

A high accuracy measurement procedure developed and validated at LGC has been transferred to a number of expert UK laboratories, and their experience in applying the technique has been evaluated by inter-laboratory comparisons. It is an “exact matching” calibration procedure for analysis of organic analytes using isotope dilution mass spectrometry (IDMS). This calibration procedure uses a calibration blend and a sample blend with closely matched isotope amount ratios, and is an iterative process, culminating in the calibration blend and sample blend having identical isotope amount ratios. It is capable of high accuracy, since systematic errors in the determination of the isotope amount ratios are cancelled out. A series of four inter-laboratory comparisons of increasing difficulty were carried out involving a number of expert laboratories. The first three comparisons used gas chromatography mass spectrometry (GC–MS) analysis of the pesticide metabolite (pp′-dichlorodiphenyl) dichloroethylene (pp′-DDE), involving both conventional calibration and IDMS exact matching procedures for pp′-DDE in a solvent and a complex liquid matrix (corn oil). The fourth comparisons utilised liquid chromatography mass spectrometry (LC–MS) and involved the analysis of sulphamethazine (4-amino-N-(4,6 dimethyl-2 pyrimidinyl) benzenesulphonamide) in solvent using IDMS and conventional calibration techniques. Following the first trial, a workshop for participants was held on the use of the exact matching procedure together with a short course on uncertainty estimation. The results of the comparisons clearly showed the superior accuracy of using IDMS with the exact matching procedure for both GC–MS and LC–MS applications. These comparisons and the workshop have enabled the methodology to be transferred to UK industry, helping to improve UK measurement capability.     

Towards compound-independent calibration for organic compounds using online isotope dilution mass spectrometry

Isotope dilution mass spectrometry (IDMS) can be considered a primary measurement method directly traceable to the International System of Units (SI). This measurement technique is increasingly employed in routine laboratories, owing to its unequalled analytical performance, precision and ease of accreditation. Unfortunately, for the adequate application of IDMS, several isotopically labelled standards, corresponding to the compounds of interest, are required. Additionally, when the enriched isotope is continuously added after a chromatographic separation, and an elemental ion source is used, it allows quantification of the different analytes being eluted from the column without requiring specific standards for each compound (online IDMS). In this article, we discuss how the traditional applicability of online IDMS for elemental speciation can be dramatically expanded by using carbon isotope tracers, oxidation or combustion reactions and a conventional molecular ion source. With such a strategy every carbon-containing compound being eluted from a chromatography system can be quantified without the need for specific standards as long as quantitative combustion/oxidation and complete elution occur. So far, only gas chromatography–combustion–mass spectrometry applications have been described, but recent results indicate the great possibilities of extending this novel approach to the quantification of organic compounds after separation by liquid chromatography.     

Routine clinical determination of lead,arsenic, cadmium,and thallium in urine and whole blood by inductively coupled plasma mass spectrometry

For the measurement of As, Cd, Pb, and Tl in urine or whole blood, judicious choices of internal standard elements for matrix correction and the development of a refined isobaric arsenic correction are necessary to produce accurate ICP-MS results. Ga and Rh are chosen as internal standards for As and Cd respectively. Bi is better for the correction of Pb and Tl than Re. An empirically derived equation relating the measurement of ¹⁶O³⁵Cl to the ⁴⁰Ar³⁵Cl contribution to the arsenic signal at mass 75 is refined by measuring the responses at mass 51 and 75 for urines with added hydrochloric acid. Overall, ICP-MS results for blood and urine are within 6% of Zeeman GFAAS results for patient samples. For surveys, the overall average of ICP-MS results is within 3% of target.     

Determination of heavy metal complexes with humic substances by HPLC/ICP-MS coupling using on-line isotope dilution technique

An isotope dilution mass spectrometric (IDMS) method has been developed for the simultaneous determination of the complexes of 11 heavy metals (Ag, Cd, Cu, Mo, Ni, Pb, Tl, U, W, Zn and Zr) with humic substances (HS) by coupling HPLC with ICP-MS and applying the on-line isotope dilution technique. The HPLC separation was carried out with size exclusion chromatography. This HPLC/ICP-IDMS method was applied to samples from a brown water, ground water, sewage and seepage water as well as for a sample containing isolated fulvic acids. The total contents of heavy metals and of their complexes were analyzed in these samples with detection limits in the range of 5–110 ng/L. The analysis of heavy metal/HS complexes from the different waters resulted in characteristic fingerprints of the distribution pattern of heavy metals in the separated HS fractions. A comparison between the total heavy metal concentrations and their portions bound to humic substances showed distinct differences for the various metals. Simultaneous ¹²C detection was used for the characterization of HS complexes not identified by UV detection and for the determination of relative DOC concentrations of chromatographic peaks. Received: 21 February 1997 / Revised: 27 May 1997 / Accepted: 28 May 1997