Toxicological evaluation of some food
additives including anticaking agents,
antimicrobials, antioxidants, emulsifiers
and thickening agents
WHO FOOD ADDITIVES SERIES NO. 5
The evaluations contained in this publication
were prepared by the Joint FAO/WHO Expert
Committee on Food Additives which met in Geneva,
25 June - 4 July 19731
World Health Organization
1 Seventeenth Report of the Joint FAO/WHO Expert Committee on
Food Additives, Wld Hlth Org. techn. Rep. Ser., 1974, No. 539;
FAO Nutrition Meetings Report Series, 1974, No. 53.
FOOD-GRADE MINERAL OIL
This substance has been evaluated for acceptable daily intake by
the Joint FAO/WHO Expert Committee on Food Additives (see Annex 1,
Ref. No. 23) in 1970.
Since the previous evaluation, additional data have become
available and are summarized and discussed in the following monograph.
The previously published monograph has been expanded and is reproduced
in its entirety below.
Mineral oils are of variable composition depending on the boiling
point of the fractions used. For food purposes usually liquid
petrolatum or liquid paraffin are employed which consist essentially
of n-alkanes and some cyclic paraffins. They are chemically inert
especially as regards the straight chain alkanes and on ingestion most
of the mineral oil (98%) remains unabsorbed in the faeces. There is
evidence now that small amounts of mineral oil (2%) are absorbed as
such by the intestinal mucosa and are distributed throughout the body.
A very small fraction may undergo further biochemical transformation.
Sources of mineral oil are laxatives or oils used in food technology
as release agents or for lubrication purposes (Boitnott & Margolis,
Oil droplets, identified as saturated alkane hydrocarbons, have
been demonstrated in mesenteric lymphnodes and nodes of the porta
hepatis in man. Similar droplets have been identified in human liver,
spleen and adipose tissue. The small amounts formed are consistent
with the calculated intake from food use (47.5 per head per year in
the United States of America). The quantities of extractable oil and
types of histological appearance have been reported (Boitnott &
Margolis, 1970). There is no correlation with age or sex (Kelsall &
Blackwell, 1969). There is an unusual geographical incidence, being
about 50% in North America and 10% in Western Europe and in post
mortem examinations (Cruikshank, 1972). No known harm appears
associated with these residues (Boitnott & Margolis, 1966).
Similar deposition of oil and minor absorption was demonstrated
in rabbits, rats and guinea-pigs fed liquid petrolatum for seven
months or more. Histochemical evidence showed absorption to be
proportionate to length of exposure. The mechanism of absorption was
unknown but the absorbed particles showed evidence of foreign body
reaction and phagocytic ingestion (Stryker, 1941).
Mineral oil used as emulsifying medium for s.c. injection was
transported from the site of injection without causing any systemic
effects (Brown, 1966).
H3-labelled mineral oil was administered to rats orally and i.p.
five hours after oral dosing with 0.66 mg/kg bw it could be shown
that over 80% was not absorbed but excreted in the faeces, 1-5%
was absorbed unchanged and another 15% appeared in carcass as H3
non-mineral oil substance. Some H3 had exchanged with available
H and possibly some mineral oil had been modified metabolically.
Radioactivity was found principally in liver, fat, kidney, brain, and
spleen. Following i.p. administration there was only very slow
excretion. 11% appeared in the faeces during eight days and only
traces in the urine (Ebert et al., 1966).
Mineral oil passes through the gut wall unchanged and more is
absorbed in the presence of powerful emulsifiers, provided that the
particle size of the emulsion is about 0.5 µ (Frazer et al., 1944).
Prolonged administration of 0.66 ml/kg for 31 days had no effect on
the amount absorbed when compared with single dosing (Ebert et al.,
A wide range of fractions of mineral oil contain carcinogenic
compounds especially higher boiling fractions of the range
300°-350°-400°C as shown by skin painting of mice and rabbits (Cook et
al., 1958) but refined material may be free from these carcinogenic
constituents (Prigal, 1967). Inoculation of 64 mice with a combination
of mineral oil and killed staphylococci induced plasma cell tumours in
seven animals (Potter & Robertson, 1961). Later experiments produced
similar results with mineral oil alone (Potter & Boyce, 1962). Some
doubts have been raised as to the probable role of virus in production
of these mouse plasma cell tumours (Prigal, 1967). No human cancer has
been reported following many years of oleothorax use (Prigal, 1967).
Work is going on on the metabolism of n-alkanes (heptadecane,
C17, eicosane, C20, - tetracosane, C24, and dotriacontane, C32), one
isoparaffin ((CH3)7nonane; C16) and one cycloparaffin
(dodecylcyclohexane; C18) in rat. The preliminary results are
summarized in the following table.
METABOLIC BALANCE OF SEVERAL ALKANES ORALLY
ADMINISTERED TO FEMALE WISTAR RAT, AT A SINGLE
DOSE OF 15 mg INCORPORATED IN THE DIET
PERCENTAGE NOT ABSORBED:
n-alkane/ (CH3)7- Dodecyl-
Days nonane cyclohexane Heptadecane Eicosane Tetracosane
(C16) (C18) (n-C17) (n-C20) (n-C24)
1 0.21 1.66 1.46 3.87 22.43
2 0.24 1.20 5.02 1.53 1.83
3 2.63 3.07 0.10 0.10 0.46
4 0 0 0 0 0.08
5 - - - - 0.02
6 - - - - 0
Total 3.08 5.93 6.58 5.50 24.82
No urine excretion was observed.
Some points have been noted:
(a) The isoparaffin is absent in the carcass.
(b) 0.8% of the ingested dose of tetracosane (C24) is retained.
(c) For the other n-alkanes, between 7 and 10% of the ingested
dose are retained after 21 days. This is equivalent to 7 ppm
(0.0007%) in the carcass.
Heptadecane and dodecylcyclohexane were given, during one week,
to eight rats at a daily dose of 15 mg incorporated in the diet. At
day 10, the animals were sacrificed. The following levels were found
in the organs or tissues: liver 2.5 ppm (0.00025%); spleen 1.5 ppm
(0.00015%); lung 2.3 ppm (0.00023%); heart 0.5 ppm (0.00005%); muscle
34 ppm (0.0034%); adipose tissue 317 ppm (0.0317%) (Tulliez & Bories,
Traces (0.001-0.1%) of n-alkanes (C17 to C31) are found normally
in plants. Protein sources from Spirula family algae contain up to
There are two possible reasons for the presence of mineral oil in
food; (1) in trace amounts from its use as a lubricant or separant
e.g. in tin-greasing before baking, or from traces on the surface of
knives used to cut dough in breadmaking, or as a coating e.g. of
fruit; (2) as a substitute for fat either because it is cheaper or in
slimming foods. The maximum daily intake is calculated to be about
100 mg of which some 80 mg are contributed from its use on the
machinery in the baking industry (Council on White Mineral Oil, 1961).
There has been a great deal of work on the effect of mineral oil
in impeding the absorption of fat soluble vitamins A (and precursors)
D. E. K. and essential fatty acids. There is no doubt that
interference with absorption can occur, particularly of carotene if
amounts in food exceed approximately 6000 ppm (Steigmann et al.,
1952). Whether the amounts likely to appear in the food of children
are of clinical importance is much less certain (assuming that it is
not used as an ingredient as in (2) above). But the diets of many of
these may contain amounts of these vitamins that are in any case
marginal or inadequate and there seems no reason for the inclusion of
mineral oil in foods which are specifically intended for infants with
the possible exception of rusks (concerning which inquiries are being
made which will be later reported as they may be subject to the same
contaminating processes as bread).
Ten rats were each fed a total of 17 g liquid paraffin in 18 g
olive oil over 16 days mixed into their normal diet. Some 65% was
absorbed as estimated from faecal loss. Another five rats received
over 28 days a total of 28 g liquid paraffin in their diet. Only 9%
was absorbed. Lymph collected during absorption from intestinal
lymphatics showed that absorbed paraffin had been metabolically
modified (Daniel et al., 1953).
Fifteen rabbits, weighing between 1.9 and 2.5 kg, were given
daily 25 ml of a mixture (1:1) of olive oil and paraffin oil (purity
not stated). The animals were sacrificed at regular intervals, after
60-406 days of treatment. At this high dosage level, from the first to
the third week, a relatively important loss of weight is noted, but
rapidly a state comparable to the controls is regained. Progressively,
the paraffin oil passes the intestinal epithelium and accumulates in
the mesenteric lymph glands, then becoming distributed in the rest of
the body, with preferential deposition in the liver and in the spleen.
Histologically, diffuse hyperplasia of reticulo endothelial cells,
somewhat similar to that seen in human Whipple's disease, is observed
Two groups of 30 mice had mineral oil applied to their skin three
times weekly at 15 mg/application for 311 and 478 days respectively.
No tumours were found (Esso Research, 1960).
Animals were kept for 15 months on diets supplemented with 10%
liquid paraffin. The liver contained 0.4% dry weight liquid paraffin.
Some active metabolism may occur but liver function was not affected
(Daniel et al., 1953). In another experiment 2% mineral oil was fed in
the diet to 30 rats for 500 days without adverse effects (Schmähl &
The natural occurrence of hydrocarbons, particularly in food of
plant origin, the use of food-grade mineral oil in food technology and
the medicinal use of this material have resulted in detectable
residues of mineral oil in human tissues. No pathological consequences
seem to follow the retention of this material. Moreover, most of the
ingested material is not absorbed but is excreted in the faeces.
Nevertheless, such tissue storage is considered to be undesirable and
exposure to food-grade mineral oil should be kept to a minimum.
There are no other recognized toxicological problems arising from
the present technological uses of food-grade mineral oil. Further work
is desirable on the elucidation of the significance of stored mineral
oil in the body. Recent technological progress results in production
of mineral oil by hydrogenation. Such oils will contain more cyclic
paraffin since aromatic components are not removed prior to
distillation. No toxicological information is available on these newer
Estimate of acceptable daily intake for man
FURTHER WORK OR INFORMATION
Elucidation of the significance of stored mineral oil in the
Boitnott, J. K. & Margolis, S. (1966) Bull. Johns Hopk. Hosp., 118,
Boitnott, J. K. & Margolis, S. (1970) Johns Hopkins Med. J., 127, 65
Borer, F. (1960) Rev. franē. études clin. et biol., 5, 47
Brown, E. A. (1966) Review of Allergy, 20, 148 & 235
Cook, J. W., Carruthers, W. & Woodhouse, D. L. (1958) Brit. med.
Bull., 14, 132
Council on White Mineral Oil (1961) Food Additive Petition 302 to US
Food and Drug Administration dated 21 February 1961
Cruikshank, B. (1972) Personal communication
Daniel, J. W. et al. (1953) Biochem. J., 54, 37
Ebert, A. G., Schleifer, C. R. & Hess, S. M. (1966) J. Pharmac. Sci.,
Esso Research (1960) Unpublished report submitted to WHO
Frazer, A. C., Schulman, J. H. & Stewart, H. C. (1944) J. Physiol,
* See relevant paragraph in the seventeenth report, pp. 10-11.
Applies only to mineral oil not made by hydrogenation process.
Kelsall & Blackwell (1969) Pathology, 1, 211
Potter, M. & Boyce, C. R. (1962) Nature, 193, 1086
Potter, M. & Robertson, J. (1961) J. nat. Cancer Inst., 25, 847
Prigal, S. J. (1967) Annals of Allergy, 25, 449
Schmähl, D. & Reiter, A. (1953) Arzneimittel-Forsch., 3, 403
Steigmann, F. et al. (1952) Gastroent., 20, 587
Stryker, W. A. (1941) Arch. Pathol., 31, 670
Tulliez, J. & Bories, G. (1973) Personal communication