Dialuminium zinc tetraoxide

Endpoint summary

Description of key information

There are no studies available in which the toxicokinetic behaviour of zinc aluminium oxide (CAS 12068 -53 -0) has been investigated.

In accordance with Annex VIII, Column 1, Item 8.8.1, of Regulation (EC) 1907/2006 and with Guidance on information requirements and chemical safety assessment Chapter R.7c: Endpoint specific guidance (ECHA, 2012), assessment of the toxicokinetic behaviour of zinc aluminium oxide is conducted to the extent that can be derived from the relevant available information. Available data on the metallic components, i.e. aluminium and zinc in their ionic form, was taken into account as supporting information.

The substance zinc aluminium oxide is an inorganic mono-constituent substance.

The substance is a white powder at 20°C and a molecular weight of 693 - 999 g/mol. No vapour pressure was determined, since the substance is not melting point up to 1000°C. The test substance does not soluble in water. A log Pow value is not available, since the substance is inorganic.

The physicochemical properties of zinc aluminium oxide suggests low oral, low inhalation and negligible dermal absorption, and thus low bioavailability. However, the substance is anticipated to release the corresponding metal ions of zinc and aluminium under strong acid conditions as present in the stomach. Whereas absorption of essential major elements like zinc is relatively high, only poor absorption is reported for aluminium.

Due to the assumed low absorption potential of zinc aluminium oxide a low distribution is expected. However, under the strong acid conditions as present in the stomach, the substance is anticipated to release the corresponding metal ions of zinc and aluminium, which are distributed widely after absorption and are mainly accumulated in the bones. Metal ions remaining in the gastrointestinal tract are likely to precipitate and therefore are not absorbed in the lower gastrointestinal tract.

No metabolism is relevant for the substance since it is an inorganic compound, which is not metabolised via physiological pathways. However, after absorption, the release of the ionic forms of zinc and aluminium due to strong acid conditions in the stomach from the parent compound is assumed in a worst case approach. The cations may interact with cell structures and physiological pathways.

Excretion of the parent substance and the unabsorbed ions of the elements are considered to be mainly via the faeces, whereas absorbed zinc and aluminium, preferably in the ionic form, may be primarily excreted via urine.

Additional information

Assessment of the Toxicokinetic Behaviour of zinc aluminium oxide

Basic toxicokinetics

In accordance with Regulation (EC) 1907/2006, Annex VIII, Column 1, Item 8.8 and with Guidance on information requirements and chemical safety assessment Chapter R.7c: Endpoint specific guidance (ECHA, 2012), assessment of the toxicokinetic behavior of zinc aluminium oxide was conducted to the extent that can be derived from the relevant available information including physicochemical and toxicological characteristics. Moreover, based on the basic assumption that the zinc and aluminium ions are the determining factors for biological activities, data on surrogate substances for zinc and aluminium compounds are considered additionally. 

Zinc aluminium oxide is a white powder which is virtually insoluble in water (Paulus, 2012) with a molecular weight of 693-999 g/mol. No signs of melting were detected up to a temperature of 1100 °C (Henke, 2011). Determination of the octanol/water partition coefficient (log Pow) of zinc aluminium oxide is not appropriate as the test substance is inorganic and does not dissolve in water or at physiological pH (7.4). However, zinc aluminium oxide may release zinc (Zn²⁺) and aluminium (Al³⁺) cations at very acidic pH conditions in the stomach.

Absorption

Absorption of a substance depends on the potential to diffuse across biological membranes, a process determined by the molecular weight, the log Pow and water solubility (ECHA, 2012).

Oral:

The smaller the molecule, the more easily it will be taken up. In general, molecular weights below 500 are favorable for oral absorption (ECHA, 2012). With a molecular weight range of 693 - 999 g/mol oral absorption of zinc aluminium oxide in the gastrointestinal tract is impeded due to the molecular size. Furthermore, as biological membranes are built as layers consisting of lipid as well as aqueous phases, passive diffusion of zinc aluminium oxide is negligible due to its insolubility in hydrophilic and lipophilic solutions.

After oral ingestion, zinc aluminium oxide may release zinc and aluminium cations in the stomach at its low pH. As the physico-chemical characteristics of the dissociation products (e.g. physical form, water solubility, molecular weight, log Pow, vapour pressure, etc.) differ from those of the parent substance prior to absorption into the blood, the predictions based upon the physico-chemical characteristics of the parent substance do no longer apply (ECHA, 2012). In general, ionized substances are considered not to readily diffuse across biological membranes (ECHA, 2012). However, absorption via the passage through aqueous pores or carriage of ionic species with the passage of water cannot be excluded for small, water-soluble molecules like zinc and aluminium ions (ECHA, 2012).

Several animal and human studies are available that evaluate absorption of zinc after ingestion of soluble and insoluble zinc components which are summarized in the EU Risk Assessment Report of zinc oxide (Risk Assessment Report Zinc oxide, 2008). Based on these data, Zn²⁺absorption is expected to take place throughout the entire small intestine with the highest rate in the jejunum (Lee et al., 1989) via passive diffusion and a carrier-mediated process (Tacnet et al., 1990) which is concentration-dependent: at low concentrations, a cysteine-rich intestinal protein (CRIP) is involved whereas binding to metallothionein seems to be involved at higher zinc concentrations (Gunshin et al., 1991; Hempe and Cousins, 1992; Sturniolo et al., 1991) especially as zinc cations induce metallothionein synthesis in intestinal mucosa cells (Richards and Cousins, 1975). The comparison of absorption of soluble and insoluble zinc salts in humans showed that bioavailability of insoluble zinc oxide accounts about 60% of that of soluble forms (Prasad et al., 1993). In general, gastrointestinal absorption of zinc cations depends on the endogenous zinc status within the body and on nutritional factors like plant proteins, alcohol and trace elements in the diet (Risk Assessment Report Zinc oxide, 2008) resulting in an absorption rate of 20 – 30 % in persons with adequate nutritional levels.

Aluminium is known to be only poorly absorbed in humans with an uptake and intestinal absorption rate of only 0.1%, reaching maximum absorption of aluminium in the serum within 1.5 – 6 hours after administration (Steinhausen et al., 2004).

However, it has to be considered that bioavailability of aluminium strongly depends on the solubility of the ingested aluminium compound: bioavailability of different aluminium compounds including aluminium citrate, aluminium chloride, aluminium nitrate; aluminium sulfate, aluminium hydroxide, finely divided aluminium metal, powdered pot electrolyte, FD&C Red 40 aluminium lake, SALP, Kasal, sodium aluminium silicate was investigated by Atomic Energy of Canada Ltd (2010). The highest fractional uptake of²⁶Al (~0.21%) was seen for aluminium sulfate and the lowest (~0.02%) for aluminium oxide with 10-fold difference between the two values. The insoluble compounds (hydroxide, oxide and powdered pot electrolyte) were less bioavailable than the soluble compounds. Moreover, bioavailability of aluminium strongly depends on the presence of dietary constituents which enhance (i.e. carboxylic acid) or inhibit (i.e. phosphate or dissolved silicate) its absorption (ATDSR, 2008).

Overall, systemic bioavailability of the parent substance is unlikely. After release of the corresponding zinc and aluminium cations, moderate absorption of zinc is expected in humans whereas absorption of aluminium is considered to be low.

Dermal:

Dermal absorption correlates with molecular size: small molecules may be taken up more easily than bigger molecules. In general, a molecular weight < 100 favors dermal absorption, whereas molecules of molecular weights > 500 may be too large (ECHA, 2012). As the molecular weight of zinc aluminium oxide ranges between 693 - 999 g/mol, dermal absorption of of zinc aluminium oxide is rather negligible. Furthermore, as a substance must be sufficiently soluble in water to diffuse from the stratum corneum into the epidermis (ECHA, 2012), dermal uptake of zinc aluminium oxide is likely to be very low. As no log Pow can be determined, QSAR analyses are technically not possible to calculate dermal absorption values. Furthermore, zinc aluminium oxide is not irritating to skin, hence, enhanced penetration due to local skin damage can be excluded. 

Overall, the low water solubility, the high molecular weight (>100) and the fact that the substance is not irritating to skin implies that dermal absorption of zinc aluminium oxide is rather unlikely.

Inhalation:

Due to the physical state of the test substance, respiratory absorption in the lung after inhalation of dust cannot be excluded. In general, particles with aerodynamic diameters < 100μm have the potential to be inhaled by humans. Particles with aerodynamic diameters < 50μm may reach the thoracic region and those < 15μm the alveolar region of the respiratory tract (ECHA, 2012). With a particle size distribution of 22 (D10) - 140 µm (D90), zinc aluminium oxide is of inhalable size and may reach the thoracic region, especially the particels of smaller diameters (Henke, 2011). However, a transfer from the respiratory tract into the blood leading to a systemic bioavailability is considered as unlikely due to the molecular weight and insolubility of the test substance which hinder transport through the epithelium and/or membranes. A systemic absorption may be possible due to mucociliary transport followed by swallowing and subsequent release and absorption of Zn²⁺and Al³⁺may occur in the gastrointestinal tract. Furthermore, deposition of inhaled particles on the respiratory surface has to be considered when endogenous clearance mechanisms are overloaded.

Overall, a systemic bioavailability of zinc aluminium oxide after inhalation of dust is likely possible due to its granulometric characteristics.

Distribution:

Distribution within the body depends on physico-chemical characteristics of a substance including molecular weight, lipophilicity and water solubility. In general, the smaller the molecule, the wider is the distribution. However, distribution of small water-soluble molecules and ions will be limited by the grade at which they diffuse across membranes (ECHA, 2012).

In regard to the physico-chemical properties of the parent substance, absorption of zinc aluminium oxide is considered as very low after oral, inhalation or dermal exposure. Thus, the assumption that zinc aluminium oxide will most probably not be distributed throughout the body seems feasible. However, zinc aluminium oxide may release Zn²⁺and Al³⁺cations in the stomach due to its low pH. 

Zinc represents an essential trace element which is necessary for divers biological processes (including molecular processes like regulation of DNA and RNA synthesis, protein and membrane metabolism, cell growth and division (Vallee and Auld, 1990); South and Summers, 1990; Berg, 1990). Therefore, zinc cations are expected to be distributed in the organism to different tissues where they are transported through cell membranes and incorporated in over 200 enzymes as cofactor and subsequently enters biochemical pathways. Several animal studies prove the distribution of Zn²⁺in nearly all tissues and tissue fluids with the highest concentrations in the small intestine followed by the kidney, liver, large intestine, lungs and spleen shortly after administration of Zn²⁺(6 hours after oral ingestion). At later time points (14 days after administration), the hair, testicles, liver and the large intestines showed the highest levels of Zn²⁺(Kossakowski and Grosicki, 1983). In general, high Zn²⁺concentrations are present in the liver, gut, kidney, skin, lung, brain, heart, pancreas, retina and sperm (Bentley and Grubb, 199; He et al., 1991; Llobet et al., 1988). In humans, Zn²⁺is absorbed in the gastrointestinal tract, bound to serum albumin and then transported to the liver and subsequently throughout the body. Plasma concentrations of approx. 1 mg/L are measured under normal conditions with a total zinc content of 1.5 – 2g in the human body (70 kg) (ATSDR, 1994). In humans, approx. 60% of total zinc is found in the muscle and about 30% are present in the bone (Wastney et al., 1986). Cleven et al. determined the highest zinc concentrations/kg tissue in the bone, hair and prostate (Cleven et al., 1993). 

The highest tissue concentrations for aluminium were found in the liver, spleen, bone, kidney (Wilhelm et al., 1990) and the brain, which contains lower aluminium concentrations than the other tissues. An equal distribution of aluminium is present within the blood between plasma and cells. Accumulation of about 50% of absorbed aluminium occurs rapidly (within 2 hours) and permanently in the skeleton, which was shown in young rats (Jouhanneau et al., 1997). Furthermore, Al is transferred to the offspring through transplacental passage and/or maternal milk (Yumoto et al., 2000).

Overall, the available information indicates that Zn²⁺and Al³⁺cations will be distributed in the organism.

Accumulation:

In general, certain metals and small ions can interact with ions in the matrix of bone. Thus, they can displace the normal constituents of the bone leading to retention of the metal or ion (ECHA, 2012). Therefore, the dissociation products of zinc aluminium oxide are considered to accumulate in bones; which was proven for Zn²⁺and Al³⁺(Cleven et al., 1993, Johanneau et al., 1997). The largest long-term deposition of aluminium occurs in the mineralization front of bones (Steinhausen et al., 2004, Yokel and McNamara, 2001).

Overall, the available information indicates that bioaccumulation of the released cations takes place in humans whereas accumulation of the parent substance is considered as negligible due to very low absorption after oral, inhalation or dermal exposure.

Metabolism

Zinc aluminium oxide is an inorganic substance and hence, will not undergo metabolic transformation.

However, after oral ingestion, zinc aluminium oxide may release Zn²⁺and Al³⁺in the stomach. Zn²⁺is bound to organic ligands and presence as a free cation in solution is rather uncommon (Gordon et al., 1981). Moreover, zinc is found in diffusible and non-diffusible forms in the blood. About 66% of the diffusible form is freely exchangeable and loosely bound to albumin (Cousins, 1985) whereas small amounts of non-diffusible zinc are bound to α2-macroglobulin in the plasma which are not freely exchangeable with other zinc ligands. Zinc is incorporated into and dissociated fromα2-macroglobulin only in the liver (Henkin, 1974).

Aluminium exists in four different forms in the living organisms: 1) as free ions, 2) low-molecular-weight complexes, 3) physically bound macromolecular complexes, and 4) covalently bound macromolecular complexes (Ganrot 1986). Thus, metabolism of aluminium is determined by its binding affinity to its ligands and finally to their metabolism. In general, aluminium forms low-molecular-weight complexes with amino acids, nucleotides and phosphates as well as with carbohydrates and organic acids. As low-molecular-weight complexes are often chelates, they may be very stable. Furthermore, complexes between aluminium and proteins, polynucleotides, and glycosaminoglycans often represent covalently bound complexes and hence are expected to be metabolically much less active than the smaller, low-molecular-weight complexes.

Excretion

In general, characteristics favorable for urinary excretion are low molecular weight, good water solubility and ionization of the molecule at pH of urine. In contrast, molecules that are excreted in the bile have a high molecular weight (ECHA, 2012). Taken these aspects into account, excretion of the unabsorbed parent substance via the bile seems to more likely that excretion via the urine due to its high molecular weight (693 - 999 g/mol) and the insolubility in water. Considering the release of Zn²⁺and Al³⁺in the gastrointestinal tract, excretion of Zn²⁺and Al³⁺has further to be taken into account. In humans, zinc is mostly excreted via the faeces (including unabsorbed dietary zinc and endogenous zinc (Spencer et al., 1976; Venugopal and Lucky, 1978; Reinhold et al., 1991; Wastney et al., 1986), whereas only 10% is eliminated with the urine (Babcock et al., 1982; Aamodt et al., 1982). Furthermore, zinc excretions via saliva, hair loss, mother milk, and sweat have been determined as minor routes (Venugopal and Lucky, 1978; Rivlin, 1983; Prasad et al., 1963; Rossowka and Nakamoto, 1992; Henkin et al., 1975).

Similar to zinc, aluminium is excreted mainly via the faeces, which was shown after repeated ingestion of aluminium chloride in mice. In general, unabsorbed aluminium is excreted via the faeces whereas absorbed aluminium is mainly excreted in the urine after years (Steinhausen et al., 2004).

In conclusion, excretion of the parent substance is considered possible via the bile. Excretion of the dissociation products Al³⁺and Zn²⁺is expected to occur mainly in the faeces.

References not included in the IUCLID:

ATSDR (Agency for Toxic Substances and Disease Registry) (1994). Toxicological profile for zinc (update). Agency for Toxic Substances and Disease Registry, Atlanta (as cited in Risk Assessment Report Zinc Oxide, 2008)

ATSDR (Agency for Toxic Substances and Disease Registry) (2008). Toxicological Profile for Aluminum. Atlanta, Department of Health and Human Services, Public Health Service.

Bentley, P.J. and Grubb, B.R. (1991). Experimental dietary hyperzincemia tissue disposition of excess zinc rabbits. Trace Elem. Med. 8, 202-207. [Cited from ATSDR, 1994, as cited in Risk Assessment Report Zinc Oxide, 2008].

Berg, J.W. (1990). Zinc finger domains, hypotheses and current knowledge. Annu. Rev. Biophys. Biophys. Chem. 19, 405-421. [Cited from Walsh et al., 1994, as cited in Risk Assessment Report Zinc Oxide, 2008].

Cleven, R.F.M.J., Janus, J.A., Annema, J.A. and Slooff, W. (1993). Integrated Criteria Document Zinc. RIVM Report No. 710401028, Bilthoven, The Netherlands (as cited in Risk Assessment Report Zinc Oxide, 2008).

Cousins, R.J. (1985). Absorption, transport, and hepatic metabolism of copper and zinc, special reference to metallothionein and ceruplasmin. Physiol. Rev. 65, 238-309. [Cited from ATSDR, 1994, as cited in Risk Assessment Report Zinc Oxide, 2008].

ECHA. 2012. Guidance on information requirements and chemical safety assessment – Chapter 7c: Endpoint specific guidance. European Chemicals Agency, Helsinki

Ganrot, P.O. (1986). Metabolism and possible health effects of aluminum. Environ Health Perspect 65:363-441.  

Gordon, E.F., Gordon, R.C. and Passal, D.B. (1981). Zinc metabolism: Basic, clinical, and behavioral aspects. J. Pediatr. 99, 341-349 (as cited in Risk Assessment Report Zinc Oxide, 2008)

Gunshin, H., Noguchi, T. and Naito, H. (1991). Effect of calcium on the zinc uptake by brush-border membrane vesicles isolated form the rat small intestine. Agric. Biol. Chem. 35, 2813-2816. [Cited from ATSDR, 1994, as cited in Risk Assessment Report Zinc Oxide, 2008].

He. L.S., Yan, X.S. and Wu, D.C. (1991). Age-dependent variation of zinc-65 metabolism in LACA mice. Int. J. Radiat. Biol. 60, 907-916. [Cited from ATSDR, 1994, as cited in Risk Assessment Report Zinc Oxide, 2008].

Hempe, J.M. and Cousins, R.J. (1992). Cysteine-rich intestinal protein and intestinal metallothionein. An inverse relationship as a conceptual model for zinc absorption in rats. J. Nutr. 122, 89-95

Henkin, R.I. (1974). Metal-albumin, amino acid interactions: Chemical and physiological interrelationships. In: Chemical and Physiological Inter Relationships in Protein-Metal Interactions Friedman M (ed.). Plenum Press, New York, NY, 299-328 (as cited in Risk Assessment Report Zinc Oxide, 2008)

Henkin, R.I., Mueller, C.W. and Wolf, R.O. (1975). Estimation of zinc concentration of parotid saliva by flameless atomic absorption spectrophotometry in normal subjects and in patients with idiopathic hypogeusia. J. Lab. Clin. Med. 86, 175-180. [Cited from ATSDR, 1994, as cited in Risk Assessment Report Zinc Oxide, 2008]

Jouhanneau, P. et al. (1997). Gastrointestinal absorption, tissue retention, and urinary excretion of dietary aluminum in rats determined by using 26Al. Clinical Chemistry 43(6): 1023-1028

Kossakowski, S. and Grosicki, A. (1983). Effect of mercuric chloride upon zinc distribution in the rat. Bull. Vet. Inst. Pulawy. 26, 67-76. [Cited from Morrisey Donohue J et al., 1994, as cited in Risk Assessment Report Zinc Oxide, 2008].

Llobet, J.M., Domingo, J.L., Colomina, M.T., Mayayo, E. and Corbella, J. (1988). Subchronic oral toxicity of zinc in rats. Bull. Environ. Contam. Toxicol. 41, 36-43 (as cited in Risk Assessment Report Zinc Oxide, 2008)

Morrisey Donohue, J., Gordon, L., Kirman, C., Roberts, W.C .and Abernathy, C. (1994). Zinc chloride and other zinc compounds. In: Hartley WR, Roberts WC and Commons BJ (eds). Drinking water health advisory: 249-305. Munitions II (as cited in Risk Assessment Report Zinc Oxide, 2008)

Prasad AS, Beck FWJ and Nowak J (1993). Comparison of absorption of five zinc preparations in humans using oral zinc tolerance test. J. Trace. Elem. Exp. Med. 6, 109-115 (as cited in EU Risk Assessment Report Zinc Oxide, May 2008)

Prasad, A.S., Schulert, A.R., Sandstead, H.H., Miale A Jr and Farid Z (1963). Zinc, iron, and nitrogen content of sweat in normal and deficient subjects. J. Lab. Clin. Med. 62, 84-89. [Cited from ATSDR, 1994, as cited in EU Risk Assessment Report Zinc Oxide, May 2008].

Reinhold, J.G., Faradji, B., Abadi, P. and Ismail-Beigi, F. (1991). Decreased absorption of calcium, magnesium, and phosphorous by humans due to increased fiber and phosphorous consumption as wheat bread. Nutr. Rev. 49, 204-206. [Cited from ATSDR, 1994, as cited in EU Risk Assessment Report Zinc Oxide, May 2008] 

Risk Assessment Report Zinc oxide (2008), Final report, European Commission – Joint Research Centre, Institute for Health and Consumer Protection, European Chemicals Bureau (ECB)

Rivlin, R.S. (1983). Misuse of hair analysis for nutritional assessment. Am. J. Med. 75, 489-493. [Cited from ATSDR, 1994, as cited in EU Risk Assessment Report Zinc Oxide, May 2008].

Rossowka, M.J. and Nakamoto, T. (1992). Caffeine decreases zinc and metallothionein levels in heart of newborn and adult rats. Pediatr. Res. 32, 330-332. [Cited from ATSDR, 1994, as cited in Risk Assessment Report Zinc Oxide, 2008].

South, T.L. and Summers, M.F. (1990). Zinc fingers. Adv. Inorg. Biochem. 8, 199-248. [Cited from Walsh et al., 1994, as cited in Risk Assessment Report Zinc Oxide, 2008].

Spencer, H., Osis, D., Kramer, L. and Norris, C. (1976). Intake, excretion, and retention of zinc in man. In: Trace Elements in Human Health and Disease. Prasad AS (ed.) Vol. 1, Zinc and copper. New York, NY. Academic Press, 345-361. [Cited from Walsh et al., 1994, as cited in Risk Assessment Report Zinc Oxide, 2008]. 

Steinhausen et al. (2004). Investigation of the aluminium biokinetics in humans: a 26Al tracer study. Food Chem Toxicol. 42(3):363-71 PMID: 14871578

Sturniolo, G.C., Montino, M.C., Rossetto, L., Martin, A., D’Inca, R., D’Odorico, A. and Naccarato, R. (1991). Inhibition of gastric acid secretion reduces zinc absorption in man. J. Am. Coll. Nutr. 10, 372-375. [Cited from ATSDR, 1994, as cited in Risk Assessment Report Zinc Oxide, 2008].

Tacnet, F., Watkins, D.W. and Ripoche, P. (1990). Studies of zinc transport into brush-border membrane vesicles isolated from pig small intestine. Biochem. Biophys. Acta 1024, 323-330. [Cited from Walsh et al., 1994, as cited in Risk Assessment Report Zinc Oxide, 2008].

Vallee, B.L. and Auld, D.S. (1990). Zinc coordination, function, and structure of zinc enzymes and other proteins. Biochemistry 29, 5647-5659. [Cited from Walsh et al., 1994, as cited in Risk Assessment Report Zinc Oxide, 2008].

Venugopal, B. and Lucky, T.D. (1978). Metal Toxicity in Mammals 2, Chemical Toxicity of Metals and Metalloids. Plenum Press, New York and London, 68-75 (as cited in Risk Assessment Report Zinc Oxide, 2008)

Walsh, C.T., Sandstead, H.H., Prasad, A.S., Newberne, P.M. and Fraker, P.J. (1994). Zinc: Health effects and research priorities for the 1990s. Environ. Health Perspect 102, 5-46 (as cited in Risk Assessment Report Zinc Oxide, 2008)

Wastney, M.E., Aamodt, R.L., Rumble, W.F. and Henkin, R.I. (1986). Kinetic analysis of zinc metabolism and its regulation in normal humans. Am. J. Physiol. 251, R398-R408. [Cited from ATSDR, 1994, as cited in Risk Assessment Report Zinc Oxide, 2008].

Wilhelm, M. et al, 1990. Aluminium Toxicokinetics. Pharmacology & Toxicology 66: 4-9

Yokel, R.A. and McNamara, P.J. (2001). Aluminium Toxicokinetics: An Updated MiniReview. Pharmacology & Toxicology 88: 159-167

Yumoto, S. et al. (2000). Transplacental passage of 26Al from pregnant rats to fetuses and 26Al transfer through maternal milk to suckling rats.Nuclear Instruments and Methods in Physics Research B 172(1-4): 925-929