JOURNALOF THE AMERICAN COLLEGE OF TOXICOLOGY

Volum'e 7, Number 6,1988

Mary Ann Liebert, Inc., Publishers

Final Report on the Safety

Assessment of Sorbic Acid and
Potassiurn Sorbate
Sorbic acid is a straight-chain monocarboxylic acid used in cosmetic formula-
tions as a preservative at concentrations up to 1.0%.
Sorbic acid and potassium sorbate were practically nontoxic to rats and
mice in acute oral toxicity studies. In subchronic studies no significant adverse
effects were observed in rats, mice, or dogs when 10% sorbic acid was
included in t h e diet.
Sorbic acid and potassium sorbate at concentrations up to 10% were
practically nonirritating to the rabbit eye. Both ingredients at concentrations
up to 10%were at most only slightly irritating.
Sorbic acid and potassium sorbate have been tested for mutagenic effects
using t h e Ames test, genetic recombination tests, reversion assays, rec assays,
tests for chromosomal aberrations, sister chromatid exchanges, and gene
mutations. Results have been both positive and negative.
Potassium sorbate at 0.1% in t h e diet or 0.3% in drinking water of rats for
up to 100 weeks produced no neoplasms. In other chronic studies, no
carcinogenic effect was demonstrated b y sorbic acid in rats or mice fed diets
containing up to 10% sorbic acid.
No teratogenic effects have been observed in pregnant mice and rats
ad ministered potassium sorbate.
In three repeat insult patch tests, sorbic acid had overall sensitization rates
of 0, 0.33, and 0.8%.All of the subjects sensitized were inducted with 20%
sorbic acid and challenged with 5% sorbic acid. Formulations containing up to
0.5% sorbic acid and or potassium sorbate were not significant primary or
cumulative irritants and not sensitizers at this test concentration. A formula-
tion containing 0.01%sorbic acid was not a photosensitizer.
On t h e basis of the available data, it i s concluded that sorbic acid and
potassium sorbate are safe as cosmetic ingredients in t h e present practices of
use (andconcentration.

INTROD UCTI 0N

T l i e literature o n sorbic acid and potassium sorbate dating from 1920 to

'1975 has been previously reviewed in a generally recognized as safe
837
838 COSMETIC INGREDIENT REVIEW

(GRAS) report and evaluation and is only briefly summarized A survey

of the most recent literature, pertinent articles not included in the GRAS
report and evaluation, and the unpublished industry data have been incorpo-
rated here.

CHEMICAL A N D PHYSICAL PROPERTIES

Sorbic acid is a straight-chain monocarboxylic acid also known as 2,4-

hexadienoic acid. Potassium sorbate is the potassium salt of sorbic acid.'3'"'
These ingredients conform to the following structures(?

H H

C C COOH

H H

SORBIC ACID

H H

H 3C c c

C C COOK

H
I H
I
POTASSIUM SORBATE

Sorbic acid i s a white, free-flowing, crystalline powder that is relatively

soluble in alcohol and ether and only slightly soluble in water. It has a faint
characteristic odor and a slightly acrid taste. Potassium sorbate is a w h i t e
crystalline p o w d e r or w h i t e granules or pellets w i t h no or slight odor. It is
soluble in alcohol and freely soluble in The physicochernical
properties of sorbic acid and potassium sorbate are presented in Table 1.
Sorbic acid occurs naturally as the lactone, parasorbic acid, in berries o f
t h e mountain ash, Sorbus aucuparia L., Rosaceae. It can b e synthesized b y
various processes, w h i c h include condensation o f crotonaldehyde and acetic
or malonic acid in pyridine solution,(6"o"1) condensation o f crotonaldehyde
and ketene in t h e presence o f boron t r i f l ~ o r i d e , ' ~ , ' ~
preparation
) f r o m I ,I ,3,5-
ASSESSMENT SORBIC ACID AND POTASSIUM SORBATE 839

TABLE 1. Physicochemical Properties

Property 5oihic acid Potassium sorbate

Apprarance White, free-flowing poivder"'~ White crystalline

Odor, taste Faint characteristic, odor,'" No or slight odor""
slightly acrid taste"'
M o l t t u l a r weight 112.13'' ' ' I 150 22" ''1
Boiling point ("C) 228 (decomposPs)'i0.'4' -
Melting point ("C) 1345'10 14.15)
270 (decomposes)'' ' "'lo

Flash point ("C) 127(10.1-iI

Ionization constant (at 25°C) 1.73 X lo-"" -
Dr risi t y (1 9/4"C) 1.204 (19/4"C)"') 1 36 (L5/20°C)"3'
Maximum absorption (chloroform) 260 nm ( € = 2300)ii"' -

PH 3.3 (020%)'~~' 8 0 (03%)"*'

Solubility ("4)
Water 0 25 (at 30°C)''0'
58.2 (at 20"C)"D'
3.8 (at loO"C)~io~
Propylene glycol 5.5 (at 2 0 " c ) " ~ J
Ethanol or methanol 12.90"0'
Ethanol, 20% 0.29ii0'
Glacial acetic acid 1 I .5""
Acetone 9 2'10'
Benzene 2.3'"'
Carbon tetrachloride 1.3'"'
Cyclohexane 0.28' lo'
Dioxane 11 0'10'
Piopanol 8.4'i0'
lsopropyl ether 2.7""
Niethyl acetate 6 l(l0'
T o I tie n e 1,9('Q
C h Io r o f orm Relatively soluble'"' Relatively insoluble'"
Ethcr Relatively soluble" Relatively insoluble'4'
Losc on drying (96 maximum) 0,5(5,13i I 01'89 I31
Residue on ignition ( O k maximum) 0.2'5.8.9 13) -
Arsenic as As (maximum) 3 ppmi5) 3 ppm"'
Lead as Pb (maximum) 10ppm,'" 20 ppm"' 10ppm ('I LO pprn@'

tetraalkoxyhexane,'") and dealkanolation and hydrolysis of a 3,s-dialkoxyhex-

anal dialkyl acetal under oxidative conditions.(4' The trans,transisomer is usu-
ally obtained and is the commercial product.('0i Potassium sorbate is prepared
by reacting sorbic acid with an equimolar portion of potassium hydroxide. The
resulting potassium sorbate may be crystallized from aqueous ethanol.(".",'O)
Numerous studies have been conducted on the stability of sorbic acid and
i t s salts. From a study on the effects of acids, heavy metal ions, and sodium
chloride on the autoxidation of sorbic acid in aqueous solution, it was
determined that acetaldehyde and fumaraldehydic acid were formed as reac-
tion products. Solutions of sorbic acid salts were stabilized against atmo-
spheric oxidation by the inclusion of gluconic acid, 8-lactone, citric acid,
EDTA, or erysorbic acid and its alkaline salts. Propyl gallate was also an
effective antioxidant for sorbic acid solutions.('2i
840 COSMETIC INGREDIENT REVIEW

McCarthy et aI.(l9) found that both temperature and type of container

affected the breakdown of sorbic acid. Aqueous solutions of sorbic acid (0.1%
w/v) stored for 12 weeks in polypropylene, polyvinyl chloride, polyethylene,
and glass containers all had significant loss on storage, except when refriger-
ated or in the presence of an antioxidant (as occurs in polyethylene-92.2%
sorbic acid remaining). The mechanism of decomposition was uncertain and
in polyvinyl chloride and glass (at 5OOC) was not linear. Although some
solutions became increasingly acidic with time, leading to improved contact
killing times, both dilution tests confirmed a loss in potency. These losses
were not always proportional to the spectrophotometric results.
Gruntova et a1.(20) also studied the stability of sorbic acid in aqueous and
polysorbate solutions; sorbic acid was oxidized more readily in the polysor-
bate solutions, with the rate influenced by the packaging material. Kondrat'eva
et a1.(21) found that the amount of sorbic acid in petrolatum and emulsified
bases stored at room temperature in metal containers started to decrease
within 1 month and reached 60-80% of the initial content of the bases. They
concluded that sorbic acid does not react with sodium lauryl sulfate or
diethylene glycol stearate. Nielsed2*)found that sorbic acid incorporated in a
cough syrup formulation did not decompose after 26 months of storage at
room temperature.
Sorbic acid formed complexes with various starches by interacting with the
amylose fraction of the starch. Sorbic acid complexed with acacia in aqueous
solution and was also absorbed by nylon and cellulose acetate. The degree of
sorbic acid uptake by nylon increased with both temperature and time and
was dependent on the pH of the solution, indicating the undissociated
molecule was the preferentially absorbed form.(12)
Shihab et al.(23) reported that urea, methylurea, ethylurea, 1,3-dimethyl-
urea, an ?,3-diethylurea increased the solubility of sorbic acid in water. The
ureas decreased the hydrophobic attraction between the acid molecules, thus
allowing the formation of hydrogen bonds between the acid and water
molecules. In another study on the solubility of sorbic acid in the presence of
12 macromolecules, it was found that the amount of solubilization was
greatest with polysorbates." 2,
Sorbic acid and potassium sorbate are analyzed primarily by chromato-
graphic techniques, including high-pressure liquid c h r ~ m a t o g r a p h y , ( ~ ~ - ~ ~ )
thin-layer chromatography (TLC),(27-29) gas-liquid c h r ~ m a t o g r a p h y , ( ~ gas
~-~~)
~ - ~ ~a ) combination of gas chromatography and mass
~ h r o m a t o g r a p h y , ( ~ and
spe~trometry.(~')Other methods of analysis include ultraviolet spectropho-
t ~ m e t r y , ( ~ colorimetry,(37)
~-~~) and an isotachophoretic separation based on
different electrophoretic mobilities.(40)Both sorbic acid and potassium sorbate
can be identified by close matching to standard infrared spectra with n o
indication of foreign materials.@)
There has been some concern in the past that sorbic acid may be
contaminated with trace amounts of its isomer, parasorbic acid (5-hydroxy-2-
hexanoic acid &lactone), which is a suspected carcinogen.(41)Stafford et aI.,('*)
using a new method combining column chromatography, thin-layer chro-
matography, and gas chromatography-mass spectral analysis, found no para-
sorbic acid in several food-grade samples of sorbic acid [method sensitive
ASSESSMENT: SORBIC ACID AND POTASSIUM SORBATE 841

down to concentrations of 20 ppm (20 mg/kg)]. Murphy and Wardle~orth('~'

described a more sensitive method in which parasorbic acid was extracted
from aqueous potassium sorbate with dichloromethane and determined by
gas chromatography using a flame ionization detector. They found no evi-
dence of parasorbic acid down to a concentration of 0.5 mg/kg in the few
samples of sorbic acid examined.

USE
Cosmetic
Sorbic acid and potassium sorbate are used in cosmetics and toiletries as
preservatives and a n t i m i ~ r o b i a l s . ( ~The
~ - ~ 1986
~ ) U.S. Food and Drug Adminis-
tration (FDA) data show that sorbic acid was used in a total of 445 products,
including primarily makeup (44%), skin care (19%), eye makeup (16%), hair
(7%), and bath (4%) preparations. Of these formulations, 62% incorporated
sorbic acid at concentrations of 50.1%; 37% incorporated sorbic acid at
concentrations of > 0.1-1%. Potassium sorbate was reported in 117 products,
prirnarily skin care (including suntan preparations) (44%), hair (34'%), and
makeup (8%) preparations. Of the formulations, 56% incorporated potassium
sorbate at concentrations of > 0.1-1%; 44% incorporated potassium sorbate at
concentrations of I 0.1%.(44)
The FDA cosmetic product formulation data presented in Table 2 are
compiled through voluntary filing of such data in accordance with Title 21 Part
720.4 (d)(l) of the Code of Federal Regulations (1979). Ingredients are listed in
prescribed concentration ranges under specific product type categories. Since
certain cosmetic ingredients are supplied by the manufacturer at less than
100% concentration, the value reported by the cosmetic formulator may not
necessarily reflect the actual concentration found in the finished product; the
actual concentration is a fraction of that reported to the FDA. Data submitted
within the framework of preset concentration ranges provide the opportunity
for overestimation of the actual concentration of an ingredient in a particular
product. An entry at the lowest end of a concentration range is considered the
same as one entered at the highest end of that range, thus introducing the
possibility of a 2- to 10-fold error in the assumed ingredient concentration.
The formulation data presented in Table 2 indicate that cosmetic products
containing sorbic acid and potassium sorbate may contact all external body
surfaces and hair, as well as ocular and vaginal mucosae. Sorbic acid addition-
ally may contact the oral mucosae. These products may be used daily or
occasionally over a period of up to several years. The frequency and length of
application can result in continuous exposure.

Noncosmet ic
Sorbic acid and potassium sorbate are effective preservatives at low con-
cenitration for the control of mold and yeast in cheese products, based goods,
fruit juices, fresh fruits and vegetables, wines, soft drinks, pickles, sauerkraut,
and certain fish and meat product^.'^) These ingredients are generally recog-
842 COSMETIC INGREDIENT REVIEW

TABLE 2. Product Formulation Data

Total no of Total no No of product formulations within

formulations containing each concentration range (%)
Product category in category ingredient > 1-5 > 0 1- 7 SO i

Sorbic acid
Baby products 55 4 i
Bubble baths and other bath 771 20 2
preparations
Eyeliner 235 12 11
Eye shadow 1406 26 24
Eye makeup rPmover 77 a 3
Mascara 325 10 6
Other eye makeup preparations 156 15 11
Fragrance preparations 1a48 10 5
Powders (dusting, face, and talcum, 759 14 7
excluding aftershave talc)
Hair conditioners, rinses, tonics I204 25 15 10
and other hair-grooming aids
Hair shampoos (noncoloring) 821 3 1 1
Hair shampoos (coloring) 27 3 3
Blushers (all types) 472 19 14 8
Foundations 472 13 2 II
Lipstick 1552 32 32
Makeup bases 462 106 1 105
Rouges 106 4 3 1
Other makeup preparations 337 21 4 17
(not eye)
Manicuring preparations 77 3 I 1
Personal cleanliness 506 6 6
products
Skin-cleansing preparations loo0 18 1 7 10
(cold creams, lotions,
liquids, and pads)
Face, body, and hand 1029 21 15 6
skin care preparations
(excluding shaving)
preparations)
Moisturizing skin care 802 23 1 13 9
preparations
Other skin care preparations 1042 22 7 15
Suntan preparations 202 7 6
~ - - -
1986 Totals 445 2 'I 68 275

Potassium sorbate
Bubble baths and other 478 4 3 1
bath preparations
Miscellaneous eye makeup 793 6 1
Hair conditioners 556 7 4
Hair shampoos (noncoloring) 538 18 14
Tonics, dressings, and 350 3 1
other hair-grooming aids
Wave sets 160 I' 2 2 10
Foundations 472 9 7 2
Skin and personal cleansing 976 7 5 2
preparations
face, body, and hand skin care 3281 51 35 16
preparations, including suntan
preparations
-
1986 Totals 117 66 51

Source: From Reference 44

ASSESSMENT; SCHBIC ACID AND POTASSIUM SORBATE 843

nized as safe direct food additives w h e n used in accordance w i t h good

manufacturing p r a c t i ~ e . ( ~ ’ ,Results
~ ~ ) o f a survey of f o o d manufacturers in 1970
indicated that the mean (weighted) level o f the addition of sorbic acid to
foods ranged from <0.01to 1.40% and that for potassium sorbate ranged
f r o m < 0.01 t o 0.58%. The Grocery Manufactures o f America has made a n
independent estimate of 0.5-0.3% for the range o f sorbate addition to food.(2’
The Joint F A O - W H O (Food and Agriculture Organization-World Health
Organization) Expert Committee on food additives has estimated the accept-
able daily intake of sorbic acid and its salts (expressed as sorbic acid) as 25
m8’, k g b o d y w e i g ht .(49)
IPotassium sorbate i s also recognized as a GRAS indirect food additive as it
migrates to f o o d from paper and paperboard products used in f o o d
packagi ng.(’’)
Sorbic acid and potassium sorbate are also used as preservatives in a
variety of p h a r m a c e ~ t i c a l s . ( l ~ * ~Th~e, ~Ophthalmic
~-~~) Advisory Review Panel
o f t h e F D A over-the-counter (OTC) drug review program has proposed that
sorblic acid used alone in concentrations of 0.1-0.2% is n o t an effective
antimicrobial agent because o f i t s limited bactericidal effects. They also
indicated that more data were required to establish the safety and effective-
ness of sorbic acid used as a preservative in combination with other approved
preservative^.(^^^^^) The OTC panel o n contraceptives and other vaginal drug
products has proposed that potassium sorbate at a concentration o f 1-3% was
safe and effective for O T C use as a vaginal douche for the relief o f minor
vaginal irritations. However, as potassium sorbate has n o t been marketed for
this purpose to a material extent in the United States, it is considered by the
FDA to b e a n e w drug within the meaning of Section 201 (p) of the Federal
Food, Drug, and Cosmetic Act [21 U.S.C. 321 (P)].(’~)
Sorbic acid and potassium sorbate also have various industrial uses. These
ingredients are both used as preservatives in starch g l ~ e . ‘ ~Sorbic ~ , ~ ~ acid
) is
used to improve the characteristics of drying oils, the gloss in alkyd type
coatings, and the milling characteristics of cold rubber.“’) Sorbic acid and
potassium sorbate are used to prevent the premature sprouting of wheat
seeds,(60,61)to dry out cut plants, such as alfalfa,(62)and in spray compositions
applied to t h e foliage of crop plants to increase yield and growth.(h3) Sorbic
acid has also been used in insect disease control. Silkworm larvae consuming
mulberry leaves that had been sprayed with 0.1-1.0% sorbic acid were pro-
tected from insect diseases caused by bacteria, fungi, and viruses.(64)

GENERAL BIOLOGY
Antimicrobial Effects
Sorbic acid and potassium sorbate have a broad spectrum of fungistatic
activity but are less active against bacteria. Their antimicrobial activity de-
pends upon the amount o f undissociated acid, w h i c h i s determined primarily
b y the dissociation constant (1.73 X l o p 5 for sorbic acid) and the pH o f the
system. O p t i m u m effectiveness i s attained at pH values up to 6.5.(3,45’
Table 3
gives the percentage of preservative undissociated related to pH value.
844 COSMETIC INGREDIENT REVIEW

TABLE 3. Percentage of Sorbic Acid Undissociated Related t o pH Value

PH % Undissociated

3 98
4 86
5 37
6 6.0
7 06

Source: From References 45 and 65

The anions of dissociated acids may be inactive owing to repulsion from

the negatively charged microbial cell wa11.(45,65) However, Eklund@@has re-
ported that the inhibition of bacteria by sorbic acid was due to both the
undissociated and dissociated acid and has calculated the effect in accordance
with a proposed mathematical model. Although the inhibitory action of the
undissociated acid was 10-600 times greater than that of the dissociated acid,
the latter was responsible for more than 50% of the growth inhibition of most
of the organisms tested at pH levels above 6.
The antimicrobial activities of sorbic acid and potassium sorbate have
been studied extensively. Bell et al.(67) tested sorbate against 66 species of
filamentous fungi, 32 species of yeast, and 6 species of lactic acid bacteria.
They reported that all organisms grew in media containing 0.1% sorbic acid at
pH 7; however, this concentration of sorbic acid inhibited the yeasts and fungi
at pH 4.5 and the bacteria at pH 3.5. Extensive tables on the antimicrobial
spectrum of sorbic acid are found in the thesis by York.(68)The minimal
inhibitory concentrations of sorbic acid for various common microbes are
given in Table 4. The reader is referred to Woodford and Adams(I2)and Sofos
and B ~ s t a @ for
~ ) more in depth reviews of the antimicrobial effectiveness of
sorbic acid and potassium sorbate.
Numerous mechanisms for microbial growth inhibition by sorbate are
found in the literature; they indicate there was little or no agreement among

TABLE 4. Effective Concentrations of Sorbic Acid Against C o m m o n Microbial5

Test organisms Minimal inhibitory concentration ( p g / m / )

( - 10' colony- forming (serial dilution test; incubation time5
units per ml) of 24 and 72 h; pH 6.0)

Staphylococcus aureus 50-1 00

Clostridiurn sporogenes 100-500
Escherichia coli 50-100
Klebsiella pneumoniae 50-1 00
Pseudornonas aeruginosa 7 00-300
Pseudornonas fluorescens 100-3oO
Pseudomonas cepacia 50- 100
Candida albicans 25-50
Saccharomyces cerevisiae 200-500
Aspergillus niger 200-500
Penicillium notatum 200-300
~ ~

Source From Reference 11

ASSESSMENT: SORBlC ACID AND POTASSIUM SORBATE 845

scientists as to the manner in which sorbic acid inhibited microorganisms.

Marly investigators have suggested that sorbic acid works by inhibiting various
enzyme systems and their reactions. Sorbic acid inhibition of sulfhydryl
enzymes, including fumarase, aspartase, succinic dehydrogenase, and yeast
alcohol dehydrogenase, has been noted.(69) Inhibition of the enzymes
enoHase,o’,69)pr~teinase,(’~)cataIase,(”) phosphopyruvic h~dratase,”~,’~) and
cytochrome c ~ x i d a s e ( ’ ~ )has also been reported. Reinhard and Radler(75)
found that high concentrations of sorbic acid did not inhibit the enzymes
aldolase, enolase, or pyruvate decarboxylase and assumed that sorbic acid
inhibited the yeast cells mainly by influencing the cell membrane and i t s
permeability. Cells of Saccharomyces ceievisiae rapidly adsorbed sorbic acid
(prirnarily the undissociated form), and disturbances of cell growth may be
caused by a reaction of sorbic acid with thiol groups of the surface of the
yeast Harada et al.(77) suggested that sorbic acid inhibited the respira-
tion of yeast through its competitive action with acetate at the site of
acetyl-CoA formation. Deak and Novak(”) suggested that interference with
active transport processes may play an important role in the inhibition of yeast
by sorbic acid. Freese et al.(79) and Sheu et a1.(80) generalized that lipophilic
acid preservatives uncouple both substrate transport and oxidative phosphory-
lation from the electron transport system. Growth was inhibited by a reduc-
tion in cellular uptake of amino acids, organic acids, phosphate, and other
c o n-Ipou nds .
It has been variously reported that sorbic acid was both effective and
ineffective as an antimicrobial in the presence of nonionic surfactants. Some
of these discrepancies have been attributed to test conditions,(’2)and it was
genierally accepted that sorbic acid was not strongly affected by the presence
of rionionic surfactants.(8’-8‘) In other interaction studies, pantothenic acid
and biotin reduced the effectiveness of sorbate against Verticillium dahliae
but thiamine did n o t F The activity of sorbic acid may also be reduced by
, ~ b ) acid can also be
interaction with or loss through the c o n t a i n e r ~ . ( ~ ~Sorbic
degraded by microbes capable of using sorbic acid as a carbon
The mode of sorbic acid degradation has been postulated as being through
de~~arboxylat ion
Potassium sorbate has synergistic antimicrobial activity with butylated
hydroxyanisole and tert-butyl hydroquinone against Staphylococcus aureus and
Salmonella t y p h i r n u r i ~ r n . (It~ ~has
) acted synergistically with sodium nitrite
and tripolyphosphate against Clostridium b o t ~ l i n u r n . ( Potassium~~) sorbate
acted synergistically with heat to inactivate four types of molds; the addition
of sucrose and sodium chloride further enhanced the inhibition of Aspergillus
The synergistic action of sorbate with sodium ~ h l o r i d e ( ~ and ’-~~)
sucrose(92)has been previously noted. A chlorine addition product of sorbic
acid was more effective than sorbic acid alone and was less affected by pH.(45)

Biochemical and Cellular Effects

Sorbic acid did not affect the protein content or the biosynthesis of RNA
and DNA in mouse embryo fibroblast cells in tissue culture.(94)
Sorbic acid (1.0 mmol/kg) and/or aminopyrine (0.4 mmol/kg) and sodium
nitrite (1.0 mmol/kg) were orally administered to groups of five rats for 3
aa COSMETIC INGREDIENT REVIEW

consecutive days. The rats were killed 24 h later and evaluated for alterations
in biochemical parameters, namely, glutamic-oxaloacetic transaminase and
glutamic-pyruvic transaminase in serum, hepatic mincrosomal drug oxidation
systems, and glucose-6-phosphate dehydrogenase and lysosomal enzymes in
hepatic soluble fractions. The simultaneous administration of aminopyrine
and sodium nitrite induced alterations in these parameters, believed to be due
to the formation of N-nitrosodimethylamine. Sorbic acid did not inhibit the
alterations produced by these chemicals, and when administered alone, did
not significantly affect these parameters.(95)
Sorbic acid strongly inhibited both the peroxidase and oxidase activity of
cabbage peroxidase and its isozymes. Sorbic acid produced a marked differ-
ence in action on the isozymes, being much more effective on isozyme I l l
than I. Inhibition was noncompetitive, and the effectiveness depended on the
concentration of sorbic acid, time of action, and pH of the medium. Inhibition
increased with decreasing pH. Sorbic acid was the most effective organic acid
assayed .(96)
Alimukhamedova and Ma~lani‘~’’reported that sorbic acid affected the
ultrastructural organization of yeast cells. Saccharornyces vini or Rhodotorula
glutinis incubated with 250-500 mg sorbic acid had an accumulation of dense
phospholipoprotein granules, numerous mitochondria of various sizes, and
vacuoles within the cells, as well as the presence of irregular nuclei.
Sorbic acid was the second active compound of 35 food ingredients
evaluated for a protective effect against cytogenetic radiation damages in the
root tip of an onion (Alliurn cepa). Sorbic acid produced a 31% reduction in
the rate of aberrant mitosis caused by 100 R irradiation.(98)
Sorbic acid combined in a 1 : I mixture with monolaurin effectively re-
duced ( > 99.9%) the viability of the 1 4 human RNA and DNA enveloped
viruses studied. The sorbic acid-monolaurin mixture was added to the cell
culture at a concentration of 1%and incubated for 1 h at 23°C. The virucidal
effect was attributed to the solubilization of the lipids and phospholipids in
the envelope by the mixture, leading to a generalized disintegration of the
viral envelope.(99)Similarly, sorbic acid enhanced the viral activity of the
nucleopolyhedrosis virus in treated larvae of the gypsy moth Lyrnantria
dispar .(IO0)
Sorbic acid had no inhibitory effect on the formation of plaque or the
development of caries in rats; however, it did enhance the activity of dex-
tranase on these factors.(lo1)

Metabolism and Excretion

The results of early metabolic studies indicated that sorbic acid was
qualitatively metabolized in the same manner as the saturated or singly
unsaturated fatty acids of the same C-atom number and was readily used as an
energy source.(”)
The metabolism of [“C] sorbic acid was studied after the administration by
stomach tube of approximately 920 mg sorbic acid per kg body weight to
female Sprague-Dawley rats. Within 4-10 h, 85% of the radioactivity was
recovered in the expired carbon dioxide, 0.4% in the feces, 2% in the urine, 3%
ASSESSMENT: SORBIC ACID AND POTASSIUM SORBATE 847

in internal organs and blood, 3% in skeletal muscles, and 6.6% in other parts of
the carcass. No radioactivity was in the liver or muscle glycogen, but some
radioactivity was associated with the lipid fraction of the carcass, internal
orgalns, and skin. The percentage of the radioactivity found in expired carbon
dioxide was independent of dosage between 61 and 1213 mg sorbic acid per
kg body weight. In similar tests, caproic acid was oxidized at the same rate
and to the same extent.'?)
Sodium sorbate or sodium caproate was administered orally to fasted
female rats at doses of 75 or 150 mg (calculated as acetone) per 100 m2of body
surface. Administration was conducted daily in two divided doses of approxi-
mately 6 g sorbic acid per kg body weight; a similar proportion was excreted
as ketone bodies. Sorbic acid and caproic acid were metabolized via acetone
bodies, and under normal conditions sorbic acid was completely oxidized to
carbon dioxide and water.(2'
In rabbits fed 3 g sorbic acid per kg body weight, the urine contained
0.1-0.2% trans,trans-muconic Small amounts of sorbic acid and mu-
conic acid also have been found in the urine of mice orally administered
aqueous solutions of sodium sorbate in doses of 40 and 3000 mg/kg body
weight. Within 4 days, 81 f10% of the sorbic acid was oxidized to carbon
dioxide and water; about 4% was found in urine, partially as muconic acid."'
The metabolism of sorbic acid was identical in animals and humans."'

ANIMAL TOXICOLOGY
Acute Toxicity
Oral
The oral LD, for sorbic acid in rats has ranged from 7.36 to 12.5 g/kg body
In rats fasted for 18 h prior to the administration of sodium
sorb'ate, the LD, (calculated as sorbic acid) was 3.6 g/kg for females and 4.3
g/kg for males. In rats that had not been fasted, the LD, was 5.9 g/kg (also
calculated as sorbic acid). The lower LD,, value for sodium sorbate compared
with that for sorbic acid was attributed to i t s more rapid absorption from the
gut.':2' The oral LD,, for sorbic acid in mice w a s greater than 8.0 g/kg body
we i j; ht .(Io3)
Verrett et aI.(lo4) also evaluated the toxicity of potassium sorbate using
emlciryonating chicken eggs. They injected up to 10.00 mg potassium sorbate
(in aqueous solution) into the air sac of the egg at 96 h incubation. The LD,,
was 2.44 mg potassium sorbate per egg.
A formulation containing 5% sorbic acid was administered by intubation at
a dose of 7.0 g/kg to five rats of unspecified strain and sex, and the rats were
observed for 7 days. One of the rats died 1 day after treatment. The four
surviving rats gained weight during the 7 day observation period.('05'
A 26 ml/kg dose of a cosmetic containing 0.15% potassium sorbate was
administered orally to five male and five female fasted Harlan-Wistar rats.
There were no signs of toxicity, and weight gains were normal during the 7
day observation period.('") Groups of five male and five female fasted
848 COSMETIC INGREDIENT REVIEW

Harlan-Fischer 344 rats were given 13 ml/kg of a bronzer(lo7) and a

moisturizer(lm) containing 0.15% potassium sorbate by gavage, and the ani-
mals were observed for 2 weeks. There were no deaths and no signs of
toxicity.

Intraperitoneal
Sparfel et aI.(lo3) reported an intraperitoneal (IP) LD, for sorbic acid in
mice of 2800 mg/kg body weight. Five mice were used for each dose.
Aqueous solutions of sorbic acid were brought to a pH of 6 with sodium
carbonate before injection of 0.5 m1/20 g. An IP LD,,value of 2820 mg/kg has
also been reported for sorbic acid in mice.(lo9)
Potassium sorbate had an IP LD, of 1300 mg/kg in mice.(’’’)

Subcutaneous
Sorbic acid had a subcutaneous LD, of 2820 mg/kg in mice.(lo9)

Short-Term to Subchronic Toxicity

Oral
Numerous studies have been conducted on the short-term to subchronic
oral toxicity of sorbic acid and potassium sorbate when administered to mice,
rats, guinea pigs, and dogs. Results have varied with dose and length of
administration.
Groups of 5-10 male and female albino rats were fed diets containing 10%
sorbic acid for 30-120 days. The test animals had a higher liver-body weight
ratio than the controls. Liver homogenates of the first-generation rats fed
sorbic acid had lower oxygen consumption than controls; homogenates of
liver from second-generation rats had a statistically significant decrease in
oxygen uptake. Feed intake and reproduction were normal.(”)
Rats fed an 8% sorbic acid diet for 90 days had no adverse effects other
than a slight enlargement of the liver. A 4% diet did not cause hepatic
enlargement. Similarly, no adverse effects were found upon histopathologic
examination of three dogs fed a diet containing 4% sorbic acid for 3 months ( 2 )
Rats and dogs fed diets containing up to 8% sorbic acid for 3 months were
not adversely affected.(’)
A diet containing 2% sorbic acid (about 2 g/kg body weight) was fed to
8-week-old Wistar rats for 10 weeks. Growth was unaffected. Livers were
slightly enlarged, although no microscopic abnormalities were noted.(2)
Groups of rats were fed diets containing 2 or 0.25% sorbic acid or
potassium sorbate for 3 months. At the 2% dose, slight increases in the
bilirubin and cholesterol content of the bile were noted as well as decreased
pancreatic chymotrypsin and amylase. Potassium sorbate also reduced the
lipase activity. At the 0.25% level, an increase was seen in the pancreatic juice
secretion, its protein content, and the activity of all its enzymes.(’) It was
concluded that the significance of these findings could not be assessed from
the data.(2)
ASSESSMENT: SORBIC ACID AND POTASSIUM SORBATE 849

,A 1%oily solution of sorbic acid administered orally to guinea pigs for 20

days produced a four- to sixfold increase in phagocytosis of staphylococcus.")
Groups of 50 male and 50 female mice were orally administered sorbic
acid at a dose of 80 mg/kg per day for 3 months. Groups of 25 male and 25
female mice were similarly administered sorbic acid at a dose of 40 mg/kg per
day or a polymeric impurity obtained from it at doses of 0, 8, or 800 mg/kg
per day for 2 months. The mice were observed for general condition and
behavior, survival, feed consumption, and weight gain. Tests were also con-
ducted to determine the effects of hunger, physical stress, and carbon tetra-
chloride poisoning on the test animals compared to controls. Those mice
administered sorbic acid for 2 months did not differ significantly from controls
in survival, feed consumption, or weight gain. Weight gain was significantly
increased in mice receiving the 800 mg/kg dose of the sorbic acid polymeric
impurity. Mice administered sorbic acid for 3 months had slightly decreased
weight gains compared to controls. Generally, the test mice reacted as well or
better than controls to conditions of stress, hunger, and administration of
carbon tetrachloride." I)
In another study, sorbic acid was administered for a period of 3 months at
a dolse of 40-80 mg/kg per day to 400 albino rats and 1900 mice. No toxic
effects on weight gain, feed consumption, or survival rate and no deleterious
effects on reactions to stress were produced. The immunobiologic activity and
detoxifying action of the liver were increased.(')
Groups of 10 rats (5 males and 5 females) were fed a diet containing 0, 'I,
2, 5, or 10% potassium sorbate for 3 months. The weight gain of the female
rats fed 5 and 10% potassium sorbate was decreased initially. Relative hepatic
weights were the same in all groups; renal weights were increased in the rats
fed 10% potassium sorbate, to a lesser degree in those fed 5% potassium
sorbate. No controls for high potassium intake were described.(2)
Potassium sorbate was administered in the diet for 3 months at concentra-
tions of 1 and 2% to two groups of eight dogs each. Weight gains were
comparable to those of the control group of four dogs. No adverse effects
attributable to potassium sorbate were found upon gross examination at
necropsy.(2)

Dermal
A dose of 2 ml/kg of a formulation containing 0.5% sorbic acid (pH not
specified) was applied by inunction 5 days/week for 4 weeks to the clipped
skin of the backs of three male and three female New Zealand albino rabbits.
Plastic collars were worn to prevent ingestion of the test material. The skin of
three of the rabbits was abraded. A control group consisted of three male and
three female rabbits. Hematologic and biochemical measurements were made
during the study. At the end of the study, the rabbits were killed, and the
internal organs were examined microscopically. No adverse effects were
produced on physical appearance, behavior, body weights, or survival, and no
systemic effects were attributed to the formulation. No gross or microscopic
lesions were produced. The intact and the abraded skin responses were the
same. Slight to moderate erythema and edema were observed in all the rabbits
during the first week, and this continued throughout the study. Slight atonia
850 COSMETIC INGREDIENT REVIEW

was observed in all the animals during the second week and continued to be
observed in four animals throughout the study. Slight desquamation was
observed in two animals during week 2 and in two other animals during week
3, and this continued until the end of the study. The skin had a mild
i nt rader ma I inf I am matory response .(I2,
A creain containing 0.15% potassium sorbate (pH not specified) was
applied daily for 90 days to the clipped backs of five male and five female
New Zealand rabbits. The cream was applied with a spatula to 10% of the total
body surface of each animal in a dose of 6 mg/cm2. Collars were worn to
prevent ingestion of the test material. There were five male and five female
control rabbits. The animals were observed for local and systemic effects. They
were killed at the end of the study, and gross and microscopic examinations
were performed. Two control rabbits and one treated rabbit died during the
study from causes not considered treatment related. Mean feed consumption,
body weights, and organ weights were normal, as were values obtained for
hematology, clinical chemistry, urinalyses, and light microscopic examination.
Incidental lesions in treated rabbits included granulomatous meningoen-
cephalitis and acute colitis. All treated animals developed slight to moderate
erythema and edema during the first week, and this continued throughout the
study. Desquamation was slight to moderate in all the rabbits. Four animals
developed fine fissures during week 3, and one animal had cutaneous fissures
and bleeding on days 46-48. Papillae were observed on the backs of two
animals during week 12. Histologically, compound-related dermatitis was
observed in 8 of the 10 treated rabbits. The dermatitis was mild and was
characterized by the presence of a few inflammatory cells in the upper dermis.
No erosion or ulceration of the dermis was observed.(l13)

Chronic Toxicity
Oral
Sorbic acid was evaluated for chronic oral toxicity in Wistar rats by
administration of 0,1.5, or 10%sorbic acid in the diet for 2 years. Experimental
groups consisted of 48 males and 48 females each, For a similar caloric intake
in all groups, a mixture of corn oil and starch (1 : I ) was added to the 0 and
1.5% sorbic acid diets at 10 and 8.5%, respectively. Body weight, feed and
water consumption, mortality, and hematologic and urinalysis parameters
were monitored. The organs of all rats were weighed and examined micro-
scopically. No changes in appearance or behavior were noted. Mortality was
similar in test and control groups. No significant effects attributable to sorbic
acid treatment were found in the hematologic and serum evaluations, urinaly-
ses, or microscopic examination. The total incidence of neoplasms (malignant
and nonmalignant), as well as the distribution of affected tissues, was not
influenced by sorbic acid treatment..The body weight gain in rats of the 1.5%
sorbic acid group did not differ significantly from that in controls; however,
rats of the 10% sorbic acid group had a statistically significant reduction in
body weight gain from weeks 26 and 39 on in the females and males,
respectively. This difference was only 5-10% of the control weight and was
ASSESSMENT: SORBIC ACID AND POTASSIUM SORBATE 851

not considered to represent a serious toxic effect because it did n o t affect

mori.ality. No consistent differences were noted in feed consumption; the
mea.n daily intake was calculated as 0.63 and 4.33 g/kg in males and 0.85 and
5.69 g/kg in females fed the diets with 1.5 and 10% sorbic acid, respectively.
M a l e rats of the high-dose group had higher thyroid gland weights and higher
thyroid gland-body weight ratios. All these animals had signs o f advanced
renal disease, and as prolonged renal damage in the rat can result in parathy-
r o i d gland hyperplasia,(l”) the increased thyroid weights were considered due
to increased parathyroid gland weights. The investigators claimed that their
evaluation was further supported since this condition was found in males
only, as glomerulonephrosis is usually less severe in females. Both males and
females of the 10% sorbic acid group had higher relative liver weights, and the
females additionally had higher relative kidney, small intestine, and ovary
weights. The higher relative liver and kidney weights were not considered
indicative o f a serious effect as they were not associated with microscopic
changes. (The livers of the high-dose females had only a marginal increase in
fatty change and focal necrosis attributed to increased intakes o f fatty acid.) In
conclusion, the investigators stated that sorbic acid had a no-effect level o f at
least 1.5% ( - 750 mg/kg per day), although the lack o f a carcinogenic effect
and the “doubtful nature” of the other effects at 10% ( - 5 g/kg per day)
indicate that the no-effect level may b e closer to 5%.(”5)
Sorbic acid was evaluated for chronic oral toxicity in a similar study in
m i c e (strain ASH/CSI) by administration of 0, I , 5, or 10% sorbic acid in the
diet for 80 weeks. Experimental groups consisted of 48 male and 50 female
m i c e each. To maintain the caloric intake in all groups, a mixture of corn oil
and starch (I : I ) was added to the 0, I, and 5% sorbic acid diets at 10, 9, and
5%, respectively. Body weights, mortality, and hematologic parameters were
monitored. At termination, the mice were killed and organs examined micro-
scopically. Organ weights were also recorded. No adverse effects attributable
to sorbic acid were noted o n mortality, hematologic parameters, or the
incidence o f lesions, including neoplasms. A statistically significant reduction
in body weight gain was noted in the males fed 5% sorbic acid and in both
males and females fed the 10% sorbic acid diet; this reduction was more
pronounced in the latter group. However, as mortality was unaffected and
these m i c e had no other adverse effects, the lower weight was considered a
”mildly unfavorable response.” Statistically significant increases were noted in
t h e relative organ weights of the brain, liver, kidney, stomach, and small
intestine o f males on both the 5 and 10% sorbic acid diets. All groups o f
females treated with sorbic acid had increased relative heart and liver weights,
and females of the highest dietary group also had increased relative brain,
small1 intestine, kidney, and spleen weights. The elevation in the relative
weights of brain, spleen, stomach, and small intestine were not considered a
toxic effect in that there were no significant differences in their absolute
weights and n o indication o f microscopic change. The increased values for
relative heart weights, occurring in females only, were not considered an
effect of sorbic acid intake. The increased relative liver weights were consid-
ered to reflect an increase in metabolic demand resulting from increased fatty
acid intake as there was a lower incidence of lesions in the livers o f mice ted
852 COSMETIC INGREDIENT REVIEW

sorbic acid than in controls. Similarly, the enlarged kidneys of these mice were
not considered a serious toxic effect in that the incidence of lesions in the
kidneys was significantly less in the treated mice than in controls. In conclu-
sion, the investigators stated that the no-effect level of sorbic acid in mice
may be considered 1%of the diet ( - 1400 g/kg per day), although because of
the nature of the effects at concentrations up to lo%, the actual no-effect
level may be substantially higher.(’”)
Sorbic acid was admin‘istered in the diet at concentrations of 0, 0.1, 0.5,
and 5.0% (0, 50, 250, and 2500 mg/kg per day) for a period of 1000 days to
groups of 50 male and 50 female rats. No differences between test and control
animals were noted in appearance, growth, mortality, or reproduction. Rats
fed through the second generation a 0.1 or 0.5% sorbic acid diet had no signs
of toxicity in respect to growth or reproduction. A group of 30 rats of the
second generation maintained on a 5% sorbic acid diet for 252 days had no
significant lesions. An unpublished report from the same laboratory described
a study in which 50 male and 50 female rats were again fed 5% sorbic acid in
the diet during their life span. Mortality was not significantly affected; the
average life span of test males compared to control males was 811 and 709
days, and the test and control females lived an average of 789 and 804 days,
respectively. No differences were reported in organ weights, and only two
neoplasms were found in each of the control and test groups. No abnormali-
ties were seen in the liver, kidneys, heart, or
Chronic oral administration of sorbic acid at concentrations of 1-500 times
the amounts used in foods had no adverse effect on the blood or internal
organs of rats, guinea pigs, rabbits, or
Shtenberg and Ignatev(’’’) studied the toxicologic effects of some combi-
nations of preservatives on both mice and rats. Groups of 25 male and 25
female mice were administered 40 mg/kg per day sorbic acid or 40 mg/kg per
day sorbic acid plus 2 mg/kg per day nisin a5 a paste prior to the main feed.
Administration continued for 17 months. A control group was fed the basal
diet only. The mice were observed for their general appearance and behavior,
feed consumption, weight gain, and survival. Organ weights were also
recorded at the end of the study. Some of these mice were tested for the
effects of physical stress (swimming with a 2 g weight on the tail) and feed
restriction. After 8 months on test, some mice from the groups receiving the
preservative combination or control diets were mated and reproduction was
studied over five generations. The test mice were given the same combination
(40 mg sorbic acid and 2 mg nisin per kg per day) from weaning to mating;
litters were monitored for weight gain for 3.5 months after weaning. Mice
receiving the sorbic acid-nisin combination had a lower survival rate than
controls. Relative weights of the liver, kidneys, and testes of mice receiving
only sorbic acid were lower than those in all other groups; however, these
were not considered adverse effects. The litters from the five-generation study
administered the sorbic acid-nisin combination gained more weight than
those receiving a benzoic acid-sodium bisulfite mixture. Those mice adminis-
tered sorbic acid or sorbic acid-nisin also had better scores on the stress tests
than those receiving benzoic acid or the benzoic acid-sodium bisulfite mix-
ture. No neoplasms were found in the control or sorbic acid-nisin groups.
ASSESSMENT: SORBIC ACID AND POTASSIUM SORBATE 853

The rats of this study similarly received 40 mg/kg per day sorbic acid
(groups of 10 males and 10 females) or 40 mg sorbic acid and 2 mg nisin per
kg per day (groups of 50 males and 50 females). Other groups received
benzoic acid and/or sodium bisulfite. Feed and water consumption, weight
gain, and hematologic parameters were monitored. The effects of stress
factors were also recorded. These consisted of feed restriction, cold stress,
centrifugation, a carbon tetrachloride detoxication test, and a renal function
test. Rats fed the sorbic acid-nisin mixture gained more weight and fared
better than those on the benzoic acid-sodium bisulfite diet under all stress
conditions except feed restriction. The results of the latter study were incon-
clusive as rats in all test groups survived longer than the controls.'"') (The
results of this study were not analyzed statistically.)
Several additive toxicity tests have been conducted with sorbic acid and
other preservatives. Ohno et a1.(''8) studied the additive toxicity of sorbic acid
and benzoic acid in groups of 20 male and 20 female Sprague-Dawley rats.
The rats were administered diets for 1 year containing concentrations of 5%
sorbic acid, 0.5% benzoic acid, 2% benzoic acid, 5% sorbic acid plus 0.5%
benzoic acid, 5% sorbic acid plus 2% benzoic acid, or a basal diet with no
supplementation. A slight growth inhibition was noted in the female rats
receiving 5% sorbic acid after 6 months; no effects were noted on the males.
No significant effects were noted in the hematologic values or in the serum
and urine analyses of the test rats when compared to controls. No distinctive
microscopic changes were noted in any experimental group. Sorbic acid and
benzoic acid did not produce additive toxicity in the rat.
E3ecause of concern about the possible contamination of sorbic acid with
parasorbic acid, two chronic oral toxicity studies were conducted in rats and
mice using sorbic acid deliberately adulterated with 1000 ppm parasorbic acid.
Groups of 48 male and 48 female Wistar rats were fed diets containing sorbic
acid or the adulterated sorbic acid at concentrations of 1.2% for 2 years.'"')
Similarly, groups of 48 male and 48 female mice were fed diets containing the
same concentrations of sorbic acid and adulterated sorbic acid for 80 weeks.('*')
The inclusion of parasorbic acid in the diet of rats had no significant effect
on feed and water consumption, weight gain, hematologic values, renal
function, serum analyses, or the incidence of lesions, including neoplasm
incidence. Mortality was slightly greater in the females of the parasorbic acid
group, but this was attributed to five rats that died or were killed between
weeks 58 and 80. No comparable difference was observed in the males. The
liver weights and relative liver weights of the females in the parasorbic acid
group were increased compared to those of the sorbic acid group; however,
this was not considered significant. The investigators concluded the sorbic
acid diet was not made more toxic by the inclusion of 1000 ppm parasorbic
''
ac id . ( 9,
The inclusion of parasorbic acid in the diet of mice produced no statisti-
cally significant effects on weight gain, hematologic values, organ weights, or
lesions, including the incidence of neoplasms. Mortality was slightly higher in
the ifemales of the parasorbic acid group (statistically significant for the last 3
weeks of the study) but was not attributable to the administration of parasor-
bic acid. Three moribund mice were killed because of severe middle ear
854 COSMETIC INGREDlENT REV1EW

infection, generalized lymphoblastoma, and papillary adenoma of the lungs;

both of the latter are common in mice. The other deaths were of differing
etiology. The females receiving sorbic acid had only an unusually low mortal-
ity rate compared with historic controls from the same laboratory, and the
mortality rate of the females receiving parasorbic acid was within the normal
range for this strain of mice. The prolonged feeding to mice of a sorbic acid
diet adulterated with lo00 ppm parasorbic acid did not lead to an increase in
the toxic effects of sorbic acid and did not have a carcinogenic effect.(’*’)

Irritation
Ocular
A modified Draize ocular irritation test was conducted to evaluate the
irritancy of sorbic acid and potassium sorbate to the rabbit eye. Sorbic acid (in
petrolatum) and potassium sorbate (in aqueous solution) were evaluated at
concentrations of 1, 5, and 10%(pH not specified). Three rabbits were used for
each dose group. Eyes were scored at 1, 2, and 24 h and daily thereafter until
all irritation had disappeared. Sorbic acid at concentrations of 1, 5, and 10%
had ocular irritation indices (at 24 h) of 0.7, 0.7, and 2 (maximum = IIO),
respectively. Potassium sorbate had an ocular irritation index of 0 at all
concentrations (at 24 h). Sorbic acid and potassium sorbate caused practically
no ocular irritation and were well tolerated under these conditions.(’2’)
A 1%aqueous solution of potassium sorbate (pH not specified) was placed
in the conjunctival sacs of one eye of each of three male and three female
rabbits of unspecified strain. The Draize irritation score, 1 day after test
material administration, was 0; the solution had no potential for eye
irritation.(122)A 0.1 ml dose of a potassium sorbate solution of unspecified
concentration was instilled into the conjunctival sac of one eye of each of six
New Zealand white rabbits. The Draize irritation score at 24 h ranged from 2
to 11, and the average of the Draize scores for 24, 48, and 72 h was 4.7. No
irritation was observed on day 7 after exposure. Some of the conjunctival
tissue in two female rabbits was bleached white on day 1, and this was also
observed in one of these rabbits on day 2. Conjunctival petechial hemorrhage
was observed in the third female on days 1-3; on day 7, this was no longer
observed.(.’23)
An eye makeup remover containing 0.10% sorbic acid was placed in one
eye of each of six albino rabbits. The Draize irritation score was 0; the
formulation was nonirritating.(’24)
A 0.1 ml dose of a cosmetic containing 0.15% potassium sorbate was
instilled into one eye of each of six albino rabbits of unspecified strain, and
the animals were observed for 7 days. Slight conjunctival redness was ob-
served 1 h after treatment, but this cleared within 24 h. Cornea and iris were
not affected.(Io6)Groups of six New Zealand albino rabbits were used to
evaluate the acute ocular irritation potential of a bronzer(lo7) and a
moi~turizer(’~ containing
~) 0.15% potassium sorbate. The undiluted formula-
tions were instilled into one eye of each rabbit in a dose of 0.1 MI, and
irritation was scored on days 1, 2, 3, and 7. Slight conjunctival hyperemia was
ASSESSMENT: SORBIC ACID AND POTASSIUM SORBATE 855

observed 1 h after treatment with both formulations, and this cleared within
24 h No other signs of irritation were observed.
C) erma1
A modified Draize irritation test was conducted to evaluate the dermal
irritancy of sorbic acid and potassium sorbate in rabbits. Sorbic acid (in
petrolatum) and potassium sorbate (in aqueous solution) were evaluated at
concentrations of 1, 5, and 10%(pH not specified). Three rabbits were used for
each dose group. Patches were applied under semiocclusive conditions. Sor-
bic acid at concentrations of 1, 5, and 10% had irritancy indices of 0, 0.2, and
0.5 (maximum = 8), respectively. Potassium sorbate had an irritancy index of 0
at all concentrations. Sorbic acid and potassium sorbate cause practically no
dermal irritation and were well tolerated under these conditions.(12')
1-heprimary skin irritation of a 1%aqueous potassium sorbate solution (pH
not specified) was evaluated with nine rabbits of unspecified strain. A single
occlusive patch was applied, and erythema and edema were scored 2 and 24 h
after removal. The primary irritation index (PII) for the test material was 0.6 of
a maximum possible of 4.0; the material was practically n ~ n i r r i t a t i n g . ' ' ~ ~ )
Sorbic acid, at a concentration of 5% in a lanoline-petrolatum paste, was
applied daily, 6 days/week for 3 weeks, to the shaved skin of three rats. The
paste was massaged lightly into the skin for 2 minutes, and the area was then
washed with water and any excess paste wiped away. Three rats received
similar applications of the lanoline-petrolatum paste without sorbic acid as
controls. Weight gain was monitored over the 3 week period. All of the rats
gained weight, those receiving petrolatum only at a rate of 5% and those
receiving sorbic acid at a rate of 3%. No irritation or other adverse effects were
rep0rt ed .( ' 0 3 )
7-he primary skin irritation of a product containing 0.5% sorbic acid (pH not
specified) was evaluated with nine rabbits of unspecified strain. Erythema and
edema were scored 2 and 24 h after a single occlusive patch was removed.
The PI1 of the test material was 0.72 of a maximum possible of 8.0; the skin
irritation potential of the material was minimal.('26)An eye makeup remover
that contained 0.10% sorbic acid (pH not specified) was evaluated for dermal
irritation with 24 h occlusive patches on the intact and abraded skin of rabbits.
The formulation did not irritate rabbit skin.(124)

Sensitization
Maurer et aI.(l2') compared the results of several methods used to assess
the c:ontact allergy of weak allergens in guinea pigs with the known epidemio-
logic data on the occurrence of hypersensitivity reactions in humans. Sorbic
acid was evaluated for sensitization by an optimization method in 10 male and
10 female Pirbright white strain guinea pigs at a concentration of 0.1% in
physiologic saline. The first week of induction consisted of intracutaneous
injections of 0.1 ml of the test solution on Monday (flank and back), Wednes-
day (back), and Friday (back). The guinea pigs were chemically depilated 2.1 h
after each injection, and the reactions were assessed 3 h later. The diameters
of the two largest erythematous reactions (in vertical alignment) and the
856 COSMETIC INGREDIENT REVIEW

skinfold thickness (as measured with a skinfold gauge) were used to deter-
mine the individual reaction volume for each animal for each reaction. For
induction weeks 2 and 3, 0.1 ml of a 1 : I mixture of sorbic acid (in saline) and
adjuvant was injected intracutaneously into the nuchal skin of each guinea pig
on Monday, Wednesday, and Friday. The first challenge was administered 2
weeks after the last induction dose. A volume of 0.1 ml of 0.1% sorbic acid in
physiologic saline was injected into a previously untreated site on the flank of
each animal. The diameter and increase in skinfold thickness of each reaction
were measured 24 h later to determine the individual reaction volumes. For
each animal, the reaction volume at challenge was compared to the mean
plus the standard deviation of the first four induction doses (considered the
individual threshold value). If the reaction volume upon challenge exceeded
the corresponding threshold value, the animal was considered sensitized. A
second epidermal challenge was administered 2 weeks after the intradermal
challenge. Occlusive patches containing 1%sorbic acid in soft white petrola-
tum were applied for 24 h to a shorn, previously untreated site. The reaction
sites were chemically depilated 21 h after patch removal and the extent of
erythema and skinfold thickness determined 3 h later. An allergic reaction was
considered a clearly discernible reddening of the reaction site. The number of
positive reactions to the first (intradermal) challenge was 4 of 20 ( P = 0.053);
the second (epidermal) challenge produced no positive reactions.

REACTIONS W I T H NITRITE

The potential formation of mutagenic or DNA-damaging reaction products

in the presence of sorbic acid or potassium sorbate and sodium nitrite has
been studied extensively. Conflicting results have been reported. High con-
centrations of sorbic acid and sodium nitrite, reacting under acidic conditions,
in most cases produced ethylnitrolic acid (ENA).(128-130) ENA has been re-
ported by some as mutagenic and a potent inhibitor of fscherichia
However, the results of an extensive study by Difate(133)estab-
c0/i.(128,131,132)
lished that nitrite, not ENA, was the mutagenic compound. He attributed the
mutagenicity of ENA reported in earlier studies to possible free nitrite contam-
ination of the test solutions.
Robach et al.,(134)
evaluating the mutagenicity of sorbic acid-sodium nitrite
reaction products produced in bacon-curing brines, reported that ENA was
not formed at nitrite concentrations < 250 ppm and at a pH of 3.4. ENA was
not formed at higher pH values (6) even with a nitrite concentration of 500
ppm. All sorbate-nitrite solutions and their ether extracts were negative in the
Ames Salmonella assays,
Osawa and Namiki(13’)analyzed the reactants of sodium nitrite with some
sorbic acid analogs for mutagenicity using the rec assay and the Ames test. By
a large-scale reaction of sodium nitrite with sorbic acid methyl ester, they
isolated and identified 5-nitro-2,4-hexadienoic acid methyl ester and ENA as
the main mutagens. The investigators concluded that a nitro group adjacent to
the double bond is an important factor for the development of mutagenicity.
ASSESSMENT SORBIC ACID AND POTASSIUM SORBATE 857

Khoudokormoff and Gist-Bro~ades(’~~)

studied the mutagenicity of several
food preservatives under conditions of pH and nitrite concentrations approxi-
mating those used in preserved foods. Bacillus subtilis mutant strain M45 rec-,
unable to repair DNA damage, was used as the test organism, with the
wilcl-type strain H17+ as control. Sorbic acid (0.1% as potassium sorbate)
solutions containing 100, 200, or 400 ppm nitrite at a pH range of 3.0-6.5
consistently had a mutagenic activity that increased as the pH decreased and
persited for 8 weeks. No mutagenicity was detected at pH 2 6. The results of
similar experiments carried out with concentrations of SO; approximating
those used in wine (150 mg/liter) were a weak mutagenic effect after 2 weeks
of exposure. Sorbic acid, nitrite, or bisulfite, tested alone at any pH, did not
have mutagenic activity. The sorbic acid-nitrite complex was not mutagenic in
two other microbiologic systems using Ames Salmonella strains and E. coli
WP:2 and WP2uvrap. Both these systems had media buffered at pH 7, a pH at
which sorbate and nitrite exert no mutagenic activity.
Namiki et a1.(136,137)studied the effects of reaction conditions on the
induced mutagenic (and antibacterial) activities of sorbic acid-sodium nitrite
reactants. The pH of the medium, relative molar ratio of nitrite to sorbic acid,
and time and temperature were all varied. Mutagenicity was assayed by the 5.
subtilis rec assay and by the Ames assay using Salmonella typhimurium TA-98
and TA-I00 strains without metabolic activation. The reaction mixture (20 mM
sorbic acid and 160 mM sodium nitrite reacted at 60°C for 1 h) obtained at a
pH above 6.0 was inactive in the rec assay. DNA-damaging activity was
produced at a pH between 2 and 5, with the maximum at 3.5-4.2. Mutagenic
activity reached a maximum at a sorbic acid-NaNO, molar ratio of 1 :8 (20
mM/l60 mM), even though the formation of mutagenic compounds was
detected at a ratio of 1 :0.5.(13’)Heating the reaction mixture to 60°C produced
maximal DNA-damaging activity within 30 minutes, which then decreased
gradually over time; the reaction carried out at 4°C had activity increasing
slowly with time and reaching a maximum level between 48 and 96 h. By use
of TLC and/or column chromatography, five C-nitro and C-nitroso com-
pounds were isolated. These compounds included ENA, product Y, deter-
mined to be 1,4-dinitro-2-methylpyrrole, product B (total structure unknown),
and products F and pre-F, considered primary products of the reaction that
would lead to secondary and tertiary products. Tested individually for muta-
genic activity, 1,4-dinitro-2-methylpyrrole was highly mutagenic by both the
rec assay and the Ames assay. ENA was highly active in the rec assay but had
no ,activity with S. typhimurium strain TA-98 and only weak activity with strain
TA-100. Product B had no mutagenicity by the Ames assay and weak activity
by the rec assay; products F and pre-F were inactive by all bioassays. Sorbic
acid (100 mM) and sodium nitrite (160-800 mM) had no mutagenicity by the
rec assay at the concentrations used in these experiments, although sodium
nitrite has had mutagenic activity in other systems.(’3h)The addition of
ascorbic acid or cysteine effectively inhibited the mutagen formation in this
reaction system.(13”
Tanaka et aI.(l3’) also tested for mutagenic activity in sorbic acid-nitrite
reaction products. A compound designated compound I (the same as product
858 COSMETIC INGREDIENT REVIEW

pre-F just discussed), as well as an unidentified product, were examined by

the Ames assay with S. typhirnuriurn strains TA-98, TA-100, and TA-1538.
Neither compoilnd gave positive results with or without metabolic activation.
These compounds also gave negative results when evaluated for DNA-damag-
ing activity using the B. subtilis rec assay. The reaction mixture itself (10 mM
sorbic acid and 100 mM NaNO,, pH 1, 37°C) was negative by the Ames assay
when evaluated without metabolic activation.
Because sorbic acid reacts readily with nitrite, it was postulated that sorbic
acid would inhibit the formation of carcinogenic nitrosamines from amines
and nitrite. Numerous investigators have studied this under varying experi-
mental conditions. Tanaka et al.(138) found that sorbic acid (20 mM) inhibited
the in vitro formation of N-nitrosodimethylamine from dimethylamine and
nitrite (40 m M NaNO,, pH 2) by up to 74%. Sorbic acid (20 mM) also inhibited
the formation of N-nitrosomorpholine from morpholine and nitrite (20 mM,
pH 2) by up to 72%. Sorbic acid had no effect on the nitrosation of N-methyl-
aniline. Ascorbic acid, tested under similar conditions, was equally inhibiting
to the formation of N-nitrosodimethylamine but was a much stronger in-
hibitor of the formation of the other nitrosamines.
Lathia and S~hellhob(’~~’ also investigated the inhibition of nitrosamine
formation in vitro by sorbic acid and/or ascorbic acid. Sorbic acid (0.05 mM)
inhibited the formation of N-methyl-N-nitrosoaniline by 54% and N-
nitrosomorpholine by 77% from N-methylaniline and morpholine, respec-
tively. Reactions were carried out at pH 2 with equimolar amounts of potas-
sium nitrite with either N-methylaniline (0.1 mM) or morpholine (10 mM).
Increasing the concentration of sorbic acid to 0.1 mM decreased the amount
of inhibition for both nitrosamines. Sorbic acid and ascorbic acid synergisti-
cally inhibited the formation of N-methyl-N-nitrosoaniline but not N-
n itrosomorphol ine.
Massey et al.(I4O)studied the effects of sorbic acid (and ascorbic acid) on
N-nitrosamine formation in a heterogeneous, protein-based model system
containing a 20% nonaqueous phase (glycerol tributyrate). The reactions were
carried out at 37°C with an aqueous phase pH of 5.25. Sorbic acid (0.05 M)
reduced the formation of nitrosopyrrolidine from pyrrolidine (0.05 M) and
sodium nitrite (0.1 M) by 50% in both the aqueous and nonaqueous phases of
the system.
Amundson et al.f141)compared the nitrosamine formation in fat and lean
bacons cured with 0.26% potassium sorbate and either 40 or 120 ppm sodium
nitrite. Nitrosamine formation was suppressed, although not eliminated, by
the sorbaie cure in both types of bacon.
Kawanishi et al.(142)reported that sorbic acid had no effect on nitrosamine
formation from either aminopyrine or arninocycline reacting with sodium
nitrite in guinea pig or rat stomachs.
Sorbic acid inhibited the formation of N-nitrosodimethylamine in human
saliva from the interaction (in vitro) of salivary nitrite with aminopyrine or
oxytetracycline. Inhibition ranged from 24 to 45% with I mM of sorbic acid
and from 51 to 81% with 10 mM sorbic acid. Inhibition was greater at pH 3
than at pH 4.(143’
ASSESSMENT: SORBIC ACID AND POTASSIUM SORBATE 859

!Potassium sorbate has been incorporated into cosmetic formulations to

minimize N-nitrosamine contamination.('"'

MUTAGENI CITY

.The results of genetic recombination tests indicated that sorbic acid had a
deleterious effect on the genetic material of B. subtilis 168. At concentrations
of 20 and 30 pg/ml, sorbic acid (pH adjusted to 7) decreased the frequency of
transformations to 77 and 75%, respectively. Concentrations of 1-10 pg/mI
sorbic acid produced at 90-91% frequency of transformation. In further test-
ing, sorbic acid (10 pg/ml) did not influence the reversion of characteristic
genetics in cells of 6. subtilis strains 3308, 112, 566, or 168. Sorbic acid (10
pg/ml) was also nonmutagenic by the Ames test with S. typhirnuriurn strains
1535 and I537.'14''
Morita et evaluated sorbic acid for mutagenicity using a rec assay
with wild and recombination-deficient strains of 13. subtilis and a reversion
assay using S. typhirnurium strains TA-98 and TA-'100, both with and without
metabolic activation. Sorbic acid was negative by both the rec assay and
reversion assay at concentrations up to 5.0 mg per disk and 10 pug per plate,
respectively. Kada(129)also reported that sorbic acid was negative in the rec
assay.
Potassium sorbate was evaluated for mutagenicity in a series of short-term
assa.ys using S. typhirnuriurn strains TA-100 and TA-98 and silkworms for
mutations, B. subtifis for rec assay (without metabolic activation, pH 5), and
hamster lung fibroblast cells for chromosomal aberrations and sister chromatid
exchanges (SCE; without metabolic activation), as well as rat bone marrow
cellis for chromosomal aberrations. Potassium sorbate was positive for chromo-
somal aberrations in hamster fibroblast cells and in the rec assay with B.
subtilis; all other results were negative. No quantitative results were given.('")
Potassium sorbate was evaluated for chromosomal aberrations and sister
chromatid exchanges in a pseudodiploid Chinese hamster cell line at concen-
trations ranging from 5 X l o p 3 to 4 X M (maximum concentration of
2X M for the SCE test). Potassium sorbate produced a significant increase
in SCE ( p = 0.05) at concentrations of 1 and 2 X lo-' M when compared with
the mean value for the saline solvent, although this was not considered a
dosage effect. However, a dose-related increase in chromosomal aberrations
was noted. The investigators concluded that potassium sorbate induced aber-
rations but did not cause a pronounced increase in SCE.('48'
lshidate et aI.(lJ9) studied the induction of chromosomal aberrations using
a Chinese hamster fibroblast cell line in vitro. No metabolic activation was
used. Potassium sorbate (in saline) at a maximum tolerated dose of 4.0 mg/ml
produced chromosomal aberrations (chromatid gaps, breaks, and transloca-
tions) in 11%of the cells within 48 h. This was considered a positive response.
Sorbic acid (in dimethylsulfoxide) was negative, producing aberrations in only
3% of the cells at a maximum tolerated dose of 1.0 mg/ml.
860 COSMETIC INGREDIENT REVIEW

lshidate et al.(149’conducted further studies using the Ames test with

S. typhimurium strains TA-92, TA-1535, TA-100, TA-1537, TA-94, and TA-98
both with and without metabolic activation. Potassium sorbate (in distilled
water) at a maximum dose of 3.0 mg per plate was negative. Sorbic acid (in
dimethylsulfoxide) at a maximum dose of 10.0 mg per plate was also negative.
Potassium sorbate was evaluated for mutagenicity in a series of microbial
assays. The results of plate tests using S. typhirnuriurn strains TA-1535, TA-1537,
and TA-1538 with a concentration of 2.5% (w/v) potassium sorbate (in
phosphate buffer, pH 7.4) were negative for reversions. Suspension tests using
the same strains of S. typhimurium and S. cerevisiae strain D4 with concentra-
tions of 2.5 and 5.0% potassium sorbate were negative both with and without
metabolic activation.(150)
Hasegawa et a1.(I5’) studied the potential of sorbic acid, potassium sorbate,
and sodium sorbate to induce chromosomal aberrations, SCE, and gene
mutations in cultured Chinese hamster V79 cells. Sorbic acid was tested at
concentrations of 350, 700, and 1050 pg/ml; potassium sorbate was evaluated
at concentrations of 5000, lO,O00, 15,000, and 20,000 pg/ml. Sorbic acid and
potassium sorbate induced chromosomal aberrations in a significant number
of cells (21 and 28%, respectively) only at the highest doses tested. The effect
of sorbic acid and potassium sorbate on SCE was very limited, although
concentration dependent, with the highest doses tested resulting in numbers
of SCE 1.2 times the control level. The increase in the numbers of SCE was
statistically significant at concentrations of 1050 pg/ml sorbic acid and at all
concentrations 2 10,000 pg/ml potassium sorbate. The same test concentra-
tions of sorbic acid and potassium sorbate produced no 6-thioguanine-
resistant mutations. The effects of change in osmotic pressure caused by the
addition of sorbic acid and potassium sorbate were also evaluated by substi-
tuting sodium chloride and potassium chloride. The induction of chromo-
somal aberrations could be partially attributed to the change in osmotic
pressure, whereas the latter did not affect the number of SCE. Sodium sorbate
was substantially more genotoxic than either sorbic acid or potassium sorbate.
Tsuchiya and Yamaha(152-’54) conducted a series of mutagenicity tests on
mice administered sorbic acid or potassium sorbate. Five groups of 77-79 male
mice were fed diets containing 0 (control), 1.34, 6.7, and 20.1% potassium
sorbate as well as 15% sorbic acid for periods of time up to 15 months. The
first test evaluated the mutagenicity of the intestinal contents of these mice.
Small and large intestinal contents were removed from 5-10 mice of all groups
at week 1 and after I, 3, and 6 months. The samples of contents taken from
mice of the same dose group at the same time were combined, homogenized,
and extracted with diethyl ether. The ether layer was evaporated, and one part
was dissolved in dimethyl sulfoxide at a concentration of 0.70 mg/ml for the
mutagenicity assay test and the other part subjected to fractionation (basic,
neutral, and acidic). Using a modified Ames assay both with and without
metabolic activation, the ether extracts of week 1 and 1 and 3 months were
nonmutagenic in S. typhirnuriurn TA-100 (extensive killing of bacteria oc-
curred at 6 months); the extracts of contents sampled up through 6 months
were nonmutagenic in S. typhirnuriurn TA-98. These tests were repeated with
strain TA-98 using the acidic, basic, and neutral components obtained by
ASSESSMENT: SORBIC ACID AND POTASSIUM SORBATE 861

fractionating the intestinal contents sampled at 3 and 6 months from the

control and sorbic acid groups only. The results with basic and neutral
components were negative. The acidic components from both samples of the
sorbic acid group were slightly mutagenic but only with metabolic activation.
The mutagenic ratios of acidic components at 3 months were slightly higher
than those at 6 months; the distribution of acidic components at 6 months
was 1.5 times greater than that at 3 months. The investigators suggested that
mutagens were gradually produced in the intestine and moved into the liver,
the site of metabolic
The urine of these mice was collected at 6 months and 100 pl samples
assayed for mutagenicity using the Ames test with S. typhimuriurn strains
TA-98 and TA-100. All results were negative. Samples (200 pl) from the control,
15% sorbic acid, and 20.1% potassium sorbate groups were also assayed using
the Ames preincubation method; results were negative. Urinary samples from
these mice treated with or without P-glucuronidase were fractionated by
XAO-2 column chromatography and assayed for mutagenicity with strain
TA-98. Only those samples from the 15% sorbic acid group were mutagenic
when metabolically activated. Mutagenic ratios were unaffected by treatment
with P-glucuronidase. Urine from mice on the 15% sorbic acid diet for 1 2
months was also collected and fractionated. These samples were nonmuta-
genic in strain TA-I00 but gave positive results in TA-98 with metabolic
activation. A comparison of the volume and pH of the urine between mice of
the control and 15% sorbic acid groups at 6 months indicated a slight increase
in volume and decrease in pH of urine of the sorbic acid g r o ~ p . ' ' ~ ~ )
Tsuchiya and Yamaha(Is4)further studied the mutagenicity of the intestinal
contents, the glutathione content in the liver, and the relative body weight-
liver weight ratios in these mice after administration of sorbic acid or potas-
siurn sorbate for 12 months. The acidic components of the intestinal contents
of 10 mice from each group were assayed for mutagenicity using 5. ty-
phirnuriurn strain TA-98 both with and without metabolic activation. Those
samples taken from the 15%sorbic acid group were mutagenic with metabolic
activation (mutagenicity increased with increasing amount in milligrams of
acidic components per plate); slight mutagenicity was noted in samples from
the 20.1% potassium sorbate group (with metabolic activation). The glu-
tathione content in the liver of the 15% sorbic acid group at 3 months was
decreased by 60% compared with that of the controls. This low concentration
was maintained for up to 1 2 months. A close correlation was noted between
the extent of depletion of hepatic glutathione content and the concentration
of sorbic acid in the diet. The relative body-liver weight ratios of animals of
the 15% sorbic acid group were clearly increased compared with those of the
other groups.

CARCINOGENICITY

Potassium sorbate was administered orally to six rats at a concentration of

0.1% in the diet and to another six rats at 0.3% in the drinking water. No
induced hepatic tumors were detected by laparotomy in these animals at 65
862 COSMETIC INGREDIENT REVIEW

weeks, and the oral administration was therefore continued for 100 weeks
(when all had died). The animals were examined postmortem, with micro-
scopic examination when appropriate; no induced tumors were found in any
of the rats.(’’)
No carcinogenic effect was demonstrated by sorbic acid in Wistar rats(”5)
or ASH/CSI mice(116)fed diets containing up to 10% sorbic acid for periods of
2 years and 80 weeks, respectively. (See Chronic Toxicity: Oral section for
more details). lshizawa et al.(155)have reported a carcinogenic effect on t h e
liver of mice fed diets containing up to 15% sorbic acid for 88 weeks.

TERATOCEN IClTY

Potassium sorbate was evaluated for teratogenicity in groups of approxi-

mately 20 pregnant mice (CD-1) and rats (Wistar-derived stock). The mice
were administered potassium sorbate as a water suspension at doses of 4.6,
21.4, 99.1, and 460.0 mg/kg body weight; the rats received doses of 3.4, 15.8,
73.3, and 340.0 mg/kg body weight. Doses were administered daily by oral
intubation o n days 6-15 of gestation. Both vehicle and positive (aspirin)
controls were used. N o significant effects were noted on nidation or on
maternal or fetal survival in either mice or rats. The number of abnormalities
seen in soft and skeletal tissues of the potassium sorbate groups did not differ
from the number occurring spontaneously in the vehicular controls.(156)

CLI NICAL IRRlTATlON A N D SENSITIZATION

Clemmensen and Hjorth(lS7)patch tested 91 dermatologic patients on the

upper back with concentrations of 0.1, 1.0, 5.0, and 10% sorbic acid and
benzoic acid in petrolatum. Occlusive patches were applied for 20 minutes
and reactions scored upon removal. Sorbic acid produced erythema in 19.8,
61.5, 64.8, and 67.4% of the patients at each of the four increasing concentra-
tions, respectively. Edema was produced by sorbic acid in 0, 1.1, 7.7, and 9.0%
of the patients at successively increasing concentrations, respectively. The
investigators noted that the positive reactions seemed to follow a dose-
response curve with a plateau at 1%.A group of 10 patients with positive
reactions to sorbic acid was selected to test the effects of local application of
an antihistamine prior to patch testing. Mepyramine (2% in water or gel) wa5
applied as a closed patch test 3 h prior to patch testing with sorbic acid.
Mepyramine produced a mean reduction in erythema1 responses of 31.4%
(range 2.2-65.6%), although in no patient was the reaction totally abolished.
Prick tests with histamine produced no reactions in these patients (See Table 5
for clinical irritation and sensitization results.)
Soschin and Leyden(158) studied the effects of sorbic acid on different body
regions when used as an ingredient in steroid preparations and other vehicles.
Patches containing a 0.1 ml sample of sorbic acid at concentrations of 0.1, 0.5,
and 1.0% in 2-isopropanol and water 1 : I were applied to the deltoid muscle,
the volar aspect of the forearm, and the upper portion of the back of 15-17
TABLE 5. C l i n i c a l Irritation a n d Sensitization

Ingredient Test method No of subjects Rcs tilts RPfwcnt

Sorbic acid
0 I % in p e t r o l a t u m Single o c c l u s i v e Dermatologic patients Erkthenia i n Etlrrna in I57
1 0% patch on u p p ~ bt a t k 91 I 8 ( I9 8%) O(0%)
5 0% (20 m i n u t e s ) 91 56 (6 I 5%) I ( I I"0)
100% 91 59 (648%) 7 (77%)
89 GO (674%) 8 (9o":,)
I nvcstigatorq noted a dose-response
curvc w i t h a plateau at I"(, \ I

In i m p r o p a n o l Singlr occl u s i v c 15- I7 High prcwalcnccl of cwtlwr t ~ r y t h c ~ n or iJ 158 >

and water patch on d e l t o i d edema a t all body i t t e s w i t h dose cs
0 I% muscle, forc?arm, r m p o n w txvidpnt bv thc intc,ciiit\ o f U
0 5% and u p p c r b a c k t h i , rcac tion, rc'actioris niost i r r t c ~ i i \ c ~ P
Z
I 0% (20 m i n u t e s ) o n the face 0
In e t h a n o l a n d water Single o c c l u s i v ~ 15
0 05% patch on c h e e k s and
0 1% forehead
0 5% (20 m i n u t e s )
L
I 0%
In p e t r o l a t u m Draire RIPTa
3
Induction
10 93 Induction w i t h '1O1%: 0 of 93 wnsitizcd 159
20 33 I n d u r t i o n w i t h 20%:1 o f 33 sensitLzPd
Challcngc, 5 Ovc,ralI wniitir,ltiori i d t i ' of 0.8";,
In p e t r o l a t u m Draize R I P 1
Induc t l o n
I O"!, I81 Induction w i t h 10% 0 o f 181 wnsitizcid 160
20% 121 Induction w i t h 20";: 1 of I 2 1 w r i i i t i z c x
C t i a I Ic ngc , 5 "0 Ovc,rall sc'nsitization ratp o f 0 33%
10%in p e t r o l a t u m Draize-Shclan5ki 50 39 totally ncTgative rc'5ponses; 1 subjec-t lb I
RlPT w i t h Z+ reartion at application 3, giver
I d a y rt'st. Jdrriiriistration c-ontit1uc.d
w i t h 0.5% sorbic acid, no further
rea c t i o i i s
0 I0% I I1 ('YC' Schwdrtr a n d Pc-c k I02 tyc' makcup r v n i o w t w a \ nonirrit~iting. I h Z
makeup remover p r o p h e t i c p a t c h test nonsensitizing, and n o n p h o t o i c n i i t i z i n g
w i t h a n d w i t h o u t IJV
Shdan\kt a n d UI1,q"" 1f1c.d Eye niakcul) rt3niovc'r L V ~ SnonirritJting, IGL
Shelanski R l P T nonsc ns it i z i n g , an c l notip h ot ose t i s i t i zi ng
C o n t r o l l c d u \ c study 54 No irritation wa\ ohwrvccl I h3
(4 w c ~ k 5 )
TABLE 5. ContinueW

Ingredient Test r n ~ t h o d No of sub/ert< R~sults Reference

0 5"6 in formation, RlPT 86 Formulation d i d not induce contact I b?

tested as a 0 5 % sensitization, minimal t o m i l d irritation
aqueous s o l u t i o n
0 2O6 in RlPT 78 Bubble bath d i d not induce allergic I65
bubble bath sensitization; minimal irritation i n
three subjects
0.2% in RIPl 52 Formulation d i d not induce allergic I66
formulation sensitization; minimal irritation
0.2%in RlPT a4 Facial conditioner did not induce I67
facial conditioner allergic sensitization; minimal to
m i l d irritation
0.2%in RIP1 98 Formulation d i d not induce alltrgic I68
formulation sensitization; minimal irritation

Potassium sorbate
015% C u m uI at ive 12 Very m i l d cumulative irritation was I69
in cream irritation test observed
0158 Cum uIat ive 12 Very m i l d cumulative irritation was 170
i n moisturizer irritation test observed
Cumulative 12 No cumulative irritation was observed 171
irritation test
0.15% Shelanski-Jordan 209-210 Formulation was not a strong irritant 172
in formulation RlPT or a strong contact sensitizer
0.15% M o d i f i e d Draize- 199-204 Bronzer was not a primary irritant 173
in bronzer Shelanski RlPT or an allergic contact sensitizer
0.1 5% M o d i f i e d Draize- 202-205 Moisturizer was not a primary irritant 174
in moisturizer Shelanski RlPT or an allergic contact sensitizer
0.1% i n facial RlPT 53 Facial scrub was a very mild cumulative I75
scrub, tested irritant but was not a primary irritant
diluted 1 :lo0 i n and d i d not induce sensitization
deionized water
RIP7 53 Facial scrub was a very mild cumulative 176
irritant but was not a primary irritant
and d i d not induce sensitization
RlPT 56 Facial scrub did not induce dermal 177
irritation or sensitization
RIPl 47 Facial scrub was not a sensitizer 178
ASSESSMENT: SORBIC ACID AND POTASSIUM SORBATE 865

subjects. The patches were occluded for 20 minutes, and sites were scored 10
minutes after patch removal using a maximum scale of 4. Of these subjects, 15
received an additional application of sorbic acid at concentrations of 0.05, 0.1,
0.5, and 1.0% on the cheeks and forehead. A high prevalence of either
erythema or edema was observed at all sites, with the dose response evident
by the intensity of the reaction. Reactions were most intense on the face; the
number of scores of 3 or 4 on the face was significantly increased compared
with the other body sites at 0.1% sorbic acid. No significant differences in the
rate or intensity of the reaction rate were noted at higher concentrations of
sorbic acid.
Soschin and Leyden"58)also investigated the effect of sorbic acid-induced
reactions on the anti-inflammatory effects of corticosteroid creams. They
compared the dermal effects of 0.1% sorbic acid in ethanol and water with the
effects of hydrocortisone cream containing 0.1% sorbic acid or potassium
sorb,ate in 17 subjects. Each was applied to the cheek and forehead without
occlusion for 20 minutes. The intensity of the reaction rate was significantly
less in the topical steroid preparations than in the ethanol-water vehicle. The
anti-inflammatory effect was not affected by sorbic acid-induced erythema.
1-he results of further studies showed that pretreatment of skin with
topical steroids to induce vasoconstriction diminished the response to sorbic
acid. Oral administration of aspirin blocked the erythematous component, and
the investigators suggested that prostaglandins were therefore important me-
diators. Systemic steroids, antihistamines, and hydroxyzine failed to influence
the erythema and edema produced by sorbic acid. In electron microscope
studies of tissue from the sites on the upper back, it was concluded that
erythema, edema, and flare in response to sorbic acid were not associated
with mast cell degranuIation.('58)
Marrulli and Maibach have conducted two Draize repeat insult patch tests
(RIPT) of sorbic acid. In each test, sorbic acid was applied in petrolatum at
concentrations of 10 and 20% during the induction period and 5% for the
challenge. The induction period consisted of ten 48 h occlusive patches (72 h
on the weekends) applied over a period of 3-5 weeks. Each patch contained
0.5 g of the test material and was applied to the same site on the lateral arm
above the elbow. Following a 2 week rest period, a challenge patch was
applied for 72 h. All reactions were scored upon patch removal on a scale of
1-4. The results of the first test were 0 in 93 sensitized when treated with 10%
sorbic acid and 1 in 33 sensitized when treated with 20% sorbic acid. This gave
an overall sensitization rate of 1 in 126, or 0.8%.('59'The results of the second
test were 0 in 181 sensitized when treated with '10% sorbic acid and 1 in 121
sensitized when treated with 20% sorbic acid. This gave an overall sensitiza-
tion rate of 1 in 302, or 0.33%. The sensitization rate of 0.8% for the subjects
treated with 20% sorbic acid w a s not statistically significant.('60)
Klauder(l6') conducted a Draize-Shelanski RIPT with sorbic acid at a
concentration of 1% in petrolatum. Closed patches containing sorbic acid
were applied at the same site every other day for a total of 12 applications. It
was not specified whether the patches were occlusive or nonocclusive. After a
2 week rest, a challenge patch was applied to the same site. Of the 50 subjects
completing the test, 49 had negative responses. A single subject had a
866 COSMETIC INGREDIENT REVIEW

2 + reaction to the third induction patch; he was given 1 day of rest and then
patched with 0.5% sorbic acid. All subsequent reactions were negative. The
investigator noted that this was probably an irritation response and was
consistent with the results of the pilot study (2 and 4% sorbic acid producing
irritation in 2 of 12 and 4 of 10 subjects, respectively).
An eye makeup remover containing 0.10% sorbic acid was tested for skin
irritation in the Schwartz and Deck prophetic patch test using 102 panelists.
+
Open and closed patches were scored on a 1 to 3 + scale, and the effect of
ultraviolet radiation was also determined. There were no reactions to open
patches. There were five I + reactions to closed patches on day 1, two
+
1 reactions on day 2, one 2 + reaction on days 1 and 2, respectively, and one
reaction on day 2 following ultraviolet radiation. The same formulation was
also tested in a Shelanski and Shelanski RlPT with an unspecified number of
panelists. There were no reactions to open patches, and there were no
+
reactions following ultraviolet radiation. There were up to three 1 reactions
to closed patches each day during the 11 days of the study. The eye makeup
remover was nonirritating, nonsensitizing, and nonphotosensitizing.(162)
An eye makeup containing 0.10% sorbic acid was tested for skin irritation
in a controlled use study with 54 panelists. No irritation was observed in any
panelist during the 4 weeks of the study. The eye makeup remover was
nonirr it at ing.( ' 63)
A formulation containing 0.504 sorbic acid was tested for skin irritation and
sensitization in an RlPT procedure using 86 panelists. The formulation was
tested as a 0.5% aqueous solution. Occlusive 24 h induction patches were
applied three times a week for 3 weeks to the upper backs of the subjects. An
untreated site was challenged with a 24 h patch during week 6 of the study.
Induction patches were each scored 24 h after removal, and the challenge
patch was scored 24 and 48 h after removal. A total of 19 panelists reacted to
induction patches: 7 had mild (pink uniform erythema covering most of the
contact site) reactions, and 12 had at most barely perceptible (minimal faint
uniform or spotty erythema) reactions. There were three barely perceptible
reactions at the 24 h challenge reading and no reactions at the 48 h reading.
The formulation under these test conditions did not induce contact sensitiza-
ti on .( ")'
A bubble bath containing 0.2% sorbic acid was tested as a 0.25% aqueous
solution for skin irritation and sensitization in an RlPT with 78 panelists. Three
occlusive 24 h induction patches were applied to the upper back of each
panelist each week for 3 weeks, and a 24 h challenge patch was applied to a
previously untreated site after a 3 weeks rest. Reactions to induction patches
were scored 24 h after patch removal, and reactions to challenge patches were
scored 24 and 48 h after patch removal. O f the 78, 3 subjects had barely
perceptible reactions to induction patches, and there were no reactions at
challenge. The bubble bath formulation did not induce allergic sensitiz-
at ion .(I 65)
An RlPT was conducted with a formulation containing 0.2% sorbic acid.
Occlusive 24 h induction patches were applied to the upper backs of 52
panelists three times a week for 3 weeks, and reactions were scored 24 h after
the removal of each patch. A 24 h challenge patch was applied to an
untreated site during week 6 of the study, and the reaction was scored 24 and
ASSESSMENT: SORBIC ACID AND POTASSIUM SORBATE 867

48 h after patch removal. Of 52 subjects, 9 had barely perceptible reactions to

at least one induction patch. Another 2 subjects had barely perceptible
reactions at one reading of the challenge site, 1 at 24 h and 1 at 48 h.
Follow-up testing was performed with the subject with the reaction at the
48 h challenge reading; no reactions were observed. The original challenge
reaction was of a nonspecific, irritant nature and was not due to allergy. Under
these test conditions, this formulation did not induce allergic sensitization.('66)
The skin irritation and sensitization of a facial conditioner containing 0.2%
sorbic acid was determined in an RlPT using 84 panelists. Three occlusive 24 h
induction patches were applied to the upper back each week for 3 weeks, and
a 24 h challenge patch was applied after a 3 week rest period. Induction
reactions were scored 24 h after patch removal, and challenge reactions were
scored 24 and 48 h after patch removal. Of 84 panelists, 23 had barely
perceptible to mild reactions to at least one induction patch. A single panelist
had a barely perceptible reaction, and another panelist had a mild reaction at
the 24 h challenge reading. There were no reactions at the 48 h challenge
reading. The facial conditioner did not induce allergic sensitization.('"'
A, formulation that contained 0.2% sorbic acid was tested in an RlPT for
skin irritation and allergic sensitization. A total of nine occlusive 24 h induc-
tion patches were applied to the upper backs of 98 panelists over a 3 week
period. Reactions to these patches were scored 24 h after patch removal.
Occlusive 24 h patches were applied to untreated sites in week 6 of the study,
and reactions to these patches were scored 24 and 48 h later. Of 98 panelists,
10 had barely perceptible to mild reactions to at least one induction patch.
There were two reactions to the challenge patch at the 24 h reading, and there
were no reactions at the 48 h reading. Under the conditions of this RIPT, this
formulation did not induce allergic sensi tization.('68)
A, white cream containing 0.15% potassium sorbate was tested for cumula-
tive irritation in 12 subjects. Each day for 21 consecutive days, 23 h occlusive
patches were applied to the backs of the subjects, and reactions to each patch
were scored every 24 h. The total composite score for the cr?:n .yas 83 of a
maximum possible score of 756, and the total score with a base of 10 subjects
was 69 of a maximum possible score of 630. There was a slight potential for
very mild cumulative irritation under the conditions of this test; the cream is
probably mildly irritating in normal use.("69)
The cumulative irritation potential of a moisturizer containing 0.15% potas-
sium sorbate was evaluated with 23 or 47 h occlusive patches applied to the
back!; of 12 subjects. Patches were applied each day, with the exception of a
holiday, for 20 consecutive days. Reactions were scored 1 h after patch
remo'val. The composite total score for the 12 subjects was 169 of a possible
maximum score of 720, and the total score with a base of 10 subjects was
140.83 of a possible maximum of 600. The moisturizer was probably a mild
irritant in normal use; there was evidence of a slight potential for very mild
cumulative irritation under the conditions of this test."70)
A moisturizer containing 0.15% potassium sorbate was evaluated for cumu-
lative irritation with 12 panelists. With the exception of a holiday, 23 or 48 h
occlusive patches were applied to the backs of the subjects each day for 20
consecutive days. Reactions were scored 1 h after patch removal. The compos-
ite total irritation score for the 12 panelists was 52 of a possible maximum of
868 COSMETIC INGREDIENT REVIEW

720, and t h e total score calculated for 10 panelists was 43.33 o f a possible
m a x i m u m of 600. There was essentially no evidence o f cumulative irritation
under the conditions o f this test.(171)
A Shelanski-Jordan RlPT was conducted w i t h a formulation containing
0.15% potassium sorbate. Occlusive induction patches were applied for 24 h
to the backs o f 209 t o 210 subjects three times a week for a total o f 10
applications. Reactions were scored at patch removal on a 0-4+ scale. A 48 h
challenge patch was applied 10-14 days later, and this reaction was scored at
patch removal. After another 7-10 days a second 48 h challenge patch was
applied, a n d this reaction was scored 48 and 72 h after patch application. A
single subject had a 2 + reaction to induction patches 9 and 10; these reactions
appeared to b e irritation d u e t o occlusive patch testing. Another subject had a
2 + reaction at the 72 h reading of the second challenge; this reaction lacked
signs o f edema. No other reactions were observed. The formulation does n o t
appear t o b e a strong irritant or a strong contact
A bronzer containing 0.15% potassium srobate was evaluated for skin
irritation a n d sensitization in a modified Draize-Shelanski RIPT. Occlusive
i n d u c t i o n patches were applied for 24 h to the upper backs or inner upper
arms of 199-204 subjects three times a week for a total of 10 applications, and
reactions were scored on a scale of 0-4+ at 24 or 48 h. To t h e same sites and
to previously unpatched sites, 48 h occlusive challenge patches were applied 3
weeks later, and these reactions were scored 48 and 72 h after application.
+
There were fourteen 1 and three 2 + reactions to induction patches. There
+
w e r e six 1 reactions to the challenge patch at the original site at the 48 h
+
reading, two 1 reactions to the challenge patch at the untreated site at the
48 h reading, five 1t reactions to the challenge patch at the original site at the
+
72 h reading, and three 1 reactions to the challenge patch at the previously
untreated site at the 72 h reading, There were no other reactions. The 1t and
2+ reactions were judged to b e irritant in nature and were not considered
clinically significant. The bronzer did n o t appear to b e a primary irritant or an
a I Ierg ic contact sens iti zer. (I 73)
A m o d i f i e d Draize-Shelanski RlPT was used to test the skin irritation and
sensitization potential of a moisturizer containing 0.15% potassium sorbate.
Occlusive induction patches were applied for 24 h three times a week for a
total of 10 applications. These patches were applied t o the upper backs or
inner upper arms o f 202-205 subjects, and reactions were scored on a scale of
0 to 4 + , 24 or 48 h later. After a 3 week rest period, 48 h occlusive challenge
patches w e r e applied to the original sites and to previously untreated sites.
These reactions were scored 48 and 72 h after patch application. There were
+
nine 1 reactions and two 2 + reactions to induction patches; these reactions
were judged irritant in nature and were n o t considered clinically significant.
+
There was o n e 1 reaction t o the challenge patches applied t o previously
untreated sites at the 48 h reading, three I + reactions to the challenge
+
patches applied to the original sites at the 72 h reading, and three 1 reactions
to t h e challenge patches applied to the previously untreated sites at the 72 h
reading. The moisturizer appeared not to b e a primary irritant or an allergic
contact sensitizer.(174)
A facial scrub containing 0.1% potassium sorbate was diluted 1 :I00 in
deionized water and was evaluated for skin irritation and sensitization in an
ASSESSMENT: SORBIC ACID AND POTASSIUM SORBATE 869

RlPT with 53 subjects. Eight semiocclusive induction patches, each 24 h in

duration, were applied to the upper arm of each of the subjects over a 2 week
period. Reactions were graded at patch removal. After a 2 week nontreatment
period, a semiocclusive challenge patch was applied to a previously untreated
site for 24 h. Reactions were graded at patch removal and 24, 48, and 72 h
later, Seven minimal erythema reactions were observed during induction, and
no reactions at any challenge reading. The facial scrub was a very mild
cumulative irritant but was not a primary irritant. The formulation did not
produce sensitization in any of the subjects tested.(’751
1-he skin irritation and sensitization of a facial scrub containing 0.1%
potassium sorbate were evaluated in an RlPT using 53 panelists. The formula-
tion was diluted 1 :I00 in deionized water. Four semiocclusive induction
patches were applied to the upper arm of each subject for 24 h each week for
2 weeks. Reactions were scored at patch removal. A 24 h semiocclusive
challenge patch was applied to a previously untreated site 3 weeks later. This
reaction was evaluated at patch removal and 24, 48, and 72 h later. There were
two minimal erythema reactions during induction and no other reactions. The
facial scrub was a very mild cumulative irritant, but it was not a primary
irritant. It did not produce sensitization in any of the subjects tested.(”6)
An RIPT was conducted using 56 panelists and a facial scrub containing
0.1% potassium sorbate. The formulation was diluted 1 :I00 by weight with
distilled water for the study. Eight 24 h semiocclusive induction patches were
applied over a 2 week period to the lateral upper arm of each subject.
Reactions were scored at patch removal. After an approximately 2 week rest
period, a 24 h semiocclusive challenge patch was applied to a previously
untreated site. Reactions to the challenge patch were graded at patch removal
and 24 and 48 h later. Two slight, transient, questionable erythema reactions
were observed during induction. No other reactions were observed during
induction or challenge. The facial scrub did not induce dermal irritation or
sen s it izat ion.(I ”1
7-he skin irritation and sensitization potential of a facial scrub containing
0.1% potassium sorbate was evaluated in an RIPT with 47 panelists. The
formulation was diluted 1 :lo0 in distilled water. Eight 24 h semiocclusive
induction patches were applied to the lateral aspect of the upper arms of the
subjects over a 2 week period, and reactions were scored on a scale of 0-5 at
patch removal. After a 2 week rest period, a 24 h semiocclusive challenge
patch was applied, and reactions were scored at patch removal and 24 and
48 h later. No reactions greater than 2 (moderate erythema) were observed
during the induction period, and no reactions at challenge were indicative of
sens i ti za t ion .(I

SUMMARY

Sorbic acid is a straight-chain rnonocarboxylic acid, also known as 2,4-

hexadienoic acid. It i s a white crystalline powder soluble in alcohol and ether
but (only slightly soluble in water. Potassium sorbate i s the potassium salt of
870 COSMETIC INGREDIENT REVIEW

sorbic acid and is a w h i t e crystalline powder or w h i t e granules or pellets freely

soluble in alcohol and water.
Sorbic acid occurs naturally as the lactone, parasorbic acid, in berries o f
t h e m o u n t a i n ash, Sorbus aucuparia L., Rosaceae. The sorbic acid used in
cosmetics is synthesized b y various commercial processes. Potassium sorbate
is prepared b y reacting sorbic acid with an equimolar portion o f potassium
hydroxide.
Solutions o f sorbic acid are subject to autoxidation and atmospheric
oxidation. Both the temperature and the type of container have also affected
t h e breakdown o f sorbic acid.
Sorbic acid and potassium sorbate are analyzed primarily b y chromato-
graphic techniques. Several analytic studies have been conducted to deter-
m i n e whether sorbic acid was contaminated with its isomer parasorbic acid, a
suspected carcinogen. No traces o f parasorbic acid were f o u n d (tests sensitive
down to a concentration of 0.5 mg/kg).
Sorbic acid and potassium sorbate are used in cosmetics and toiletries as
preservatives and antimicrobials generally at concentrations o f I 1%. Accord-
i n g to t h e data voluntarily reported to the FDA through 1986, sorbic acid and
potassium sorbate were used in 445 and 117 cosmetic formulations, respec-
tively. These ingredients are primarily used in facial and eye makeup and skin
care a n d hair preparations.
Sorbic acid and potassium sorbate are generally recognized as safe (GRAS)
direct f o o d additives. They are used as preservatives at low concentrations
(<0.01-1.40%) in many foods. Potassium sorbate is also a GRAS indirect f o o d
additive as it migrates to f o o d from paper products used in packaging.
The Joint Food and Agricultural Organization-World Health Organization
Expert Committee o n food additives has estimated the acceptable daily intake
of sorbic acid and i t s salts (expressed as sorbic acid) as 25 mg/kg body weight.
Sorbic acid and potassium sorbate are used as preservatives in a variety of
pharmaceuticals. These chemicals also have various industrial uses.
Sorbic acid and potassium sorbate have a broad spectrum of fungistatic
activity b u t are less active against bacteria. Their antimicrobial activity de-
pends upon the amount of undissociated acid, w h i c h in turn is determined
primarily b y the dissociation constant and the pH o f the system. O p t i m u m
effectiveness is attained at pH values up t o 6.5. The mechanism by w h i c h
sorbic acid inhibits microorganisms i s not yet understood.
In biochemical studies, sorbic acid did not affect the protein content or
t h e biosynthesis o f RNA and DNA in mouse embryo fibroblast cells. Sorbic
acid did not significantly affect biochemical parameters w h e n administered
orally to rats. Sorbic acid did inhibit both peroxidase and oxidase activity in
cabbage a n d reduced the rate o f aberrant mitosis caused b y irradiation in
onion root tips. Sorbic acid also affected the ultrastructural organization o f
yeast cells and effectively reduced the viability of the 14 human RNA and
D N A enveloped viruses w h e n combined with monolaurin.
The results of metabolic studies were that sorbic acid was qualitatively
metabolized in the same manner as the saturated or singly unsaturated fatty
acids of t h e same C-atom number. Under normal conditions, sorbic acid was
almost completely oxidized to carbon dioxide and water.
ASSESSMENT: SORBIC ACID AND POTASSIUM SORBATE 871

Sorbic acid and potassium sorbate were practically nontoxic to rats and
mice in acute oral toxicity studies. lntraperitoneal LO, values in mice were
2800 and 2820 mg/kg for sorbic acid and 1300 mg/kg for potassium sorbate.
Sorbic acid had a subcutaneous LD,? of 2820 mg/kg in mice. Formulations
containing up to 5% sorbic acid administered orally at doses up to 7.0 g/kg
were not toxic to rats.
In short-term to subchronic oral studies, sorbic acid did not produce
significant adverse effects in rats, mice, or dogs at concentrations up to 10%
(of ithe diet). Potassium sorbate was practically nontoxic in rats and dogs at
concentrations up to I0 and 2%, respectively. Application to rabbit skin of
formulations containing 0.5% sorbic acid or 0.15% potassium sorbate over
short-term and subchronic periods, respectively, resulted in dermatitis.
Chronic oral studies in which sorbic acid was administered to mice and
rats at concentrations up to 10% have established absolute no-effect levels of
1.5% in rats and 1.0%in mice. N o significant toxic effects were noted in rats at
a 5% concentration in the diet. Sorbic acid had no additive toxicity in rats
when administered with benzoic acid. Adulteration of 2 1.2% sorbic acid diet
with 1000 ppm parasorbic acid produced not adverse effects in rats or mice
administered these diets for 2 years and 80 weeks, respectively.
Ijorbic acid (in petrolatum) and potassium sorbate (as aqueous solution) at
concentrations of 1, 5, and 10%were practically nonirritating and nonirritating,
respectively, to the rabbit eye. Formulations containing 0.1% sorbic acid or
0.15% potassium sorbate were nonirritating to the rabbit eye.
tsorbic acid (in petrolatum) and potassium sorbate (as aqueous solution) at
concentrations of 1, 5, and 10% were slightly irritating and nonirritating,
respectively, when evaluated using a modified Draize irritation test. In another
Draize test, sorbic acid was classified a severe irritant after application of 1 mg
under occlusive conditions. A 1% aqueous potassium sorbate solution was
practically nonirritating to rabbit skin. No irritation or adverse effects were
produced in rats by daily application, 6 days/week for 3 weeks, of 5% sorbic
acid in a lanoline-petrolatum paste. A formulation containing 0.5% sorbic acid
was not irritating to rabbit skin.
In a guinea pig sensitization test, sorbic acid produced four positive
reactions to the first intradermal challenge although the reactions of all 20
guinea pigs were negative at the second epidermal challenge.
The results of studies of the potential formation of mutagenic or DNA-
damaging reaction products in the presence of sorbic acid or potassium
sorbate and sodium nitrite have varied. Sorbic acid and sodium nitrite, when
reacted under acidic conditions, produced ethylnitrolic acid, considered by
some to be mutagenic. Other reaction products, only partially identified, were
both mutagenic and nonmutagenic. On the other hand, sorbic acid, in that it
reacts readily with nitrite, has inhibited the formation of some carcinogenic
nitrosamines from amines and nitrites.
Ijorbic acid and potassium sorbate have been extensively tested for muta-
geniic effects using the Ames test, genetic recombination tests, reversion
assays, rec assays, and tests for chromosomal aberrations, sister chromatid
exchanges, and gene mutations. These tests have been conducted in various
systems: B . subtilis strains 3308, 112, 566, and 168; S. typhirnuriurn strains
872 COSMETIC INGREDIENT REVIEW

TA-98, TA-100, TA-1535, TA-1537, and TA-1538; S. cerevisiae strain D4; silk-
worms; Chinese hamster cells; and rat bone marrow cells. The results have
been both positive and negative.
A series of mutagenicity tests has also been used t o evaluate the intestinal
contents and urine of mice fed sorbic acid and potassium sorbate for periods
of up t o 1 5 months. The concentration of glutathione in the liver and the
relative body weight-liver weight ratios were evaluated as well. Acidic com-
ponents of the intestinal contents and the urine of those mice administered a
diet containing 15% sorbic acid were mutagenic in S. typhirnuriurn strain
TA-98, but only with metabolic activation. The concentration of lipid peroxide
i n the livers increased almost linearly with the concentration of sorbic acid in
the diet. Sorbic acid decreased hepatic glutathione concentrations and in-
creased the relative body weight-liver weight ratios in these mice.
The oral administration of potassium sorbate as 0.1% of the diet or 0.3% of
the drinking water for up t o 100 weeks produced no neoplasms in rats. No
carcinogenic effect was demonstrated by sorbic acid in rats or mice fed diets
containing u p t o 10% sorbic acid for periods of 2 years and 80 weeks,
respectively. A diet containing up t o 15% sorbic acid has been reported to
have a carcinogenic effect i n the liver of mice after 88 weeks’ administration.
N o teratogenic effects have been observed i n pregnant mice and rats
administered potassium sorbate at doses of up t o 460 and 340 mg/kg body
weight, respectively.
In three repeat insult patch tests using a total of 478 subjects, sorbic acid
had overall sensitization rates of 0, 0.33, and 0.8%. All the subjects sensitized
were inducted with 20% sorbic acid and challenged with 5% sorbic acid.
Formulations containing u p to 0.5% sorbic acid or 0.15% potassium sorbate
were not cumulative irritants or were very mild cumulative irritants. They were
not primary irritants and were not sensitizers. A formulation containing 0.01%
sorbic acid was not a photosensitizer.

CONCLUSION

On the basis of the data included in this report, the CIR Expert Panel
concludes that sorbic acid and potassium sorbate are safe as cosmetic ingredi-
ents i n the present practices of use and concentration.

ACKNOWLEDGMENT

Elizabeth Meerman Santos, Senior Scientific Analyst, prepared the litera-

ture review for this report.

REFERENCES

1 . FOOD A N D DRUG RESEARCH LABS (FDRL). (June 1973). Scientific Literature Reviews on Generally
Recognized as Safe (CRAS) Food Ingredients-Sorbic Acid and Derivatives. Prepared for the FDA.
National Technical Information Service report No. PB-223-864.
ASSESSMENT: SORBIC ACID AND POTASSIUM SORBATE

2 the Health Aspects of Sorbic Acid and Its Salts as Food Ingredients. Prepared for the FDA, Contract N o .
FDA 223-75-2004. Life Sciences Research Office, FASEB, Bethesda, M D . NTlS Document No. PB-262
663
3 CHICHESTER, D.F., and TANNER, F.W., JR. (1972). Antimicrobial food additives In. Furia. T.E (cd ).
CRC Handbook of Food Additives, 2nd ed., Vol. 1 . Cleveland: CRC Press, pp. 115-81.
4 OSOC, A. (ed ). (1980). Remington’s Pharmaceutic-a/Scienc.es, 16th ed. Easton, PA: Mack Publishing.
5 FOOD CHEMICALS CODEX (FCC). (1981). Food and Nutrition Board, National Research Council.
Washington, D C.. National Academy Press.
6 COSMETIC, TOILETRY A N D FRAGRANCE ASSOCIATION (CTFA). (May 13, 1986) Submission of
unpublished data by CTFA. Cosmetic Ingredient Chemical Description for Sorbic Acid. CTFA No. 3-1- I.
Available for review: Director, Cosmetic Ingredient Review, I I10 Vermont Ave, N . W , Suite 810,
Washington, D.C. 20005.
7 CTFA (May 13, 1986). Submission of unpublished data by CTFA. Cosmetic Ingredient Chemical
Description for Potassium Sorbate. CTFA No. 3-1-2. Available for review. Director, Cosmetic lngredicnt
Review, 1110 Vermont Avr., N.W., Suite 810, Washington, D.C. 20005.
8 ESTRIN, N.F., HAYNES, C.R., and WHELAN, J.M. (eds ) (1982). CTFA Specihcationr/Spe~tra Washing-
ton, D.C.. Cosmetic, Toiletry and Fragrance Association.
9 JAPAN COSMETIC INDUSTRY ASSOCIATION (JCIA). (1979). lapanesc Standards of Cosmetic Ingrrdi-
~ n t s Tokyo:
. Yakuji Nippo. Ltd.
10 WINDHOLZ, M. (ed.). (1983) The Merck Index, 10th ed. Rahway, NJ: Merck and Co.
11 WALLHAUSER, K.H. (1984). Antimicrobial preservatives u w d by the cosmetic industry. I n Kabara, 1.1.
(ed.). Cosmetic and Drug Preservation-Princi~les and Practfce, New York: Marcel Dekker, pp
605-745.
12 WOODFORD, R., and ADAMS, E (1970). Sorbic acid. Am Perfum Cosmet 85(3), 25-30,
13 THE NATIONAL FORMULARY (NF). (1979) 15th ed RockLTille, M D The United States Pharmacopeial
Convention
14 SAX, N.I. (1979). Dangerous Properties of Industrial M a t e r d s , 5th Pd N e w York: Van Nostrand
Reinhold.
15 WEAST, R C (ed.). (1982). CRC Handbook of Chemistry and Physics, 63rd ed. Boca Raton, FL: CRC
Press.
16 TOXICOLOGY DATA BANK FILE. (1984). National Library of Medicine Online Services. Bethesda, M D .
17 GRASSO, P., GANGOLLI, S D., and HOOSEN, J. (1969). Connective tissue response to a short-term
series of subcutaneous injections of sorbic acid or aflatoxin. Phvsico-chemical factors determining
reaction to sorbic acid. Br. J. Cancer 23(4), 787-99
18 DICKENS, F.. JONES, H.E., and WAYNFORTH, H.B. (1968). Further tests on the carcinogenicity ot sorbic
acid in the rat. Br. j. Cancer 22(4), 762-8.
19 MCCARTHY, T.J., CLARKE, C.R., and MYBURGH, J A. (May 1973) Antibacterial effectiveness of stored
sorbic acid solutions. Cosmet. Perfum. 88, 13-6.
20 GRUNTOVA, Z , PALENIKOVA, Z., and NOZDROVICZI, P. (1978). Sorbic acid a5 a preservative for
oil-water emulsions. Farm. Obz. 47(9), 387-99.
21 KONDRAT’EVA, T.S., IVANOVA, L.A., and KOREN‘KOVA, N.N. (1975) Stability of sorbic acid in
emulsion bases for oil-water ointments (Russian). Farmatsiya (Moscow) 24(6), 71-5
22 NIELSEN, A. (1981). Stability testing of bromhexine and sorbic acid in cough syrup. Arch. Pharm. Chem
Sci. Educ. 9(6), 143-8.
23 SHIHAB, FA., COSCROVE, F.P., and ISAACSON, E.I. (1979) Studics on the solubility of sorbic acid in
aqueous solutions of urea and urea-N-alkyl derivatives. Pharmazie 34(12), 824-6.
24 AITZETMUELLER, K., and ARZBERGER, E (1984). Analysis of prcservatives in fatty foods by HPLC
(German). Z. Lebensm. Unters. Forsch. 178(4), 279-83.
25. STIJVE, T I and HISCHENHUBER, C (1984). High performance liquid chromatographic detprmination of
l o w levels of benzoic acid and sorbic acid in yogurts. Dtsch. Lebensm. Rundsch 80(3), 81-4
26. MCCALLA, M A . , MARK, F C ,and KIPP, W.H. (Jan. 1977) High performance liquid chromatographic
determination of sorbic acid in wine. J Assoc. Off Anal. Chem 60(1), 71-2.
27 YANC, Z. (1983). Determination of sorbic acid and benzoic acid in foods (Chinese). Zhongguo
Niangzao. 2(3), 32-4
28. TIAN, G.H., and KONTER, T. (Nov 1972). Thin-layer chromatographic identification of preservatives i n
food. J. Assoc. Off Anal. Chem. 55, 1223-5.
874 COSMETIC INGREDIENT REVIEW

29. MANDROU, 6.. and BRESSOLLE, F. (May 1980). Thin-layer chromatography reflectometric determina-
tion of benzoic and sorbic acids i n fruit beverages. J. Assoc. Off. Anal. Chem. 63(3), 675-8.
30. ZYGMUNT, L C (July 1979). Gas-liquid chromatographic determination of sorbic acid and sodium
benzoate i n table syrup. J . Assoc. Off. Anal. Chem. 62(4), 939-42.
31. LARSSON, B.K. (1983). Gas-liquid chromatographic determination of benzoic acid and sorbic acid In
foods: NMKL (Nordic Committee on Food Analysis) collaborative study. J. Assoc Off. Anal Chem.
66(3), 775-80.
32. COELHO, R.G., and NELSON, D.L. (1983). Rapid extraction and gas-liquid chromatographic determina-
tion of benzoic and sorbic acids in beverages. 1. Assoc. Off. Anal. Chem. 66(1), 209-11.
33 LOPEZ, J., and SIMAL, I.(1982). Determination of sorbic and benzoic acids in foods by gas chromatog-
raphy (Spanish). An. Bromatol. 34(1), 123-31.
34. WATANABE, T., and MUROTA, S. (1981). Gas chromatographic analysis of some food additives
(Japanese). Kobe Joshi Daigaku Kiyo 12, 81-6.
35. CAO, H., YANC, Z., WEI, R., and SONG, F. (1982). Determination of sorbic and benzoic acids i n food
(Chinese). Shipin. Yu. Faxiao. Gongye. 6, 19-25.
36. BOLIN, H.R., STAFFORD, A.E., and FLATH, R.A. (1984). Increased specificity in sorbic acid determina-
tion in stored dried prunes. J. Agric. Food Chem. 32(3), 683-5.
37. STAFFORD, A.E. (1976). Rapid analysis of potassium sorbate in dried prunes by ultraviolet or colorimet-
ric procedures. I.Agric. Food Chem. 24(4), 894-5.
38. LOPEZ, J., and SIMAL, J. (1982). Determination of sorbic and benzoic acids by second derivative UV
spectroscopy (Spanish). An. Bromatol. 34(1), 113-21,
39. ALMELA, L., and LOPEZ-ROCA, J.M. (1984). Simultaneous determination of sorbic and benzoic acids by
differential ultraviolet spectrophotometry. Application to fruit juices (French). Sci. Aliment. 4(1), 37-41
40 RUBACH, K., BREYER, C., and KIRCHHOFF, E. (Feb. 1980). lsotachophoretic determination of preserva-
tives (German) (author’s translation). Z. Lebensm. Unters. Forsch. 170(2), 99-102.
41. DICKENS, F., and JONES, H.E. (1963). Further studies on the carcinogenic and growth inhibiting activity
of lactones and related substance. Br. J. Cancer 19, 392-403.
42. STAFFORD, A.E., BLACK, D . R , HADDON, W.F., and WAISS, A.C., JR. (1972). Analysis and improved
synthesis of parasorbic acid. J. Sci. Food Agric. 23(6), 771-6.
43. MURPHY, J.M., and WARDLEWORTH, D.F. (1973). Improved method for the estimation of parasorbic
acid i n sorbic acid. J. Sci. Food Agric. 24(2), 253-5.
44. FDA. (1986). Product formulation data. Computer printout. Washington, D.C.
45. WILKINSON, J.B., and MOORE, R.J. (eds.). (1982). Harry’s Cosmeticology, 7th ed. New York. Chpmical
Publishing Co.
46. MCCARTHY, T.J. (1984). Formulated factors affecting the activity of preservatives. In: Kabara, J J. (ed ).
Cosmetic Drug and Preservation- Principles and Practice, New York: Marcel Dekker, pp. 359-88.
47. CODE OF FEDERAL REGULATIONS (CFR). (April 1, 1984). Title 21, Part 182.3089.
48. CFR. (April 1, 1984). Title 21, Part 182.3640.
49. F A O / W H O JOINT EXPERT COMMITTEE O N FOOD ADDITIVES. (1974). Sorbic acid and its calcium,
potassium and sodium salts. In: Toxicological Evaluation of Some Food Additives lncluding Anticaking
Agents, Antimicrobial, Antioxidants, Emulsifiers a n d Thickening Agents. W H O Food Additive Series,
N o . 5. Geneva, Switzerland: World Health Organization, pp. 121-7.
50. CFR. (April 1, 1984). Title 21, Part 182.90
51. ROSSOFF, 1.5. (1974). Handbook of Veterinary Drugs. New York- Springer Publishing.
52. KOTANI, Y., and HASEGAWA, M . (April 6, 1978). Sorbic acid composition. Ger, Offen. Patent No.
2744243 (Nippon Synthetic Chemical Industry, Co., Ltd.).
53. CHUGAI PHARMACEUTICAL CO., LTD. (Feb. 25, 1983). Anticancer pharmaceuticals containing vita-
m i n D derivatives (Japanese). Jpn. Kokai Tokkyo Koho Patent No. 83 32823.
54. PEAT, R.F. (March 27, 1984). Treatment of progesterone deficiency and related conditions with a stable
composition of progesterone and tocopherols. U.S. Patent N o 4439432.
55. FOOD A N D DRUG ADMINISTRATION (FDA). (May 6, 1980). Establishment of a monograph and
proposed rulemaking on ophthalmic drug products for Over-The-Counter human use. Fed. Reg. 45(89),
30002-50
56. FDA. (June 28, 1983). Tentative final monograph for ophthalmic drug products for over-the-counter
human use. Fed. Reg. 48(125), 29788-800.
57. FDA. (Oct. 13, 1983). Establishment of a monograph for vaginal drug products for over-the-counter
human use. Fed. Reg. 48(199), 46694-729.
ASSESSMENT: SORBIC ACID AND POTASSIUM SORBATE 875

58. HOKKO CHEMICAL INDUSTRY CO , LTD (Oct 23, 1980) Starch glue preservatives (Japanese) Jpn
Kokai Tokkyo Koho Patent No 80136208
59. NIPPON SYNTHETIC CHEMICAL INDUSTRY CO LTD (Feb 23 1981) Industrial preservative's Ipn
Kokal Tokkyo Koho Patent N o 81 18903
60. OVCHAROV K E MURSHOVA N D and KOSHELEV Y P (Frb 5 1977) Protecting freshlv collected
wheat seeds against premature \prouting U S S R Patpnt N o 515314 (TimiryareL K A Institute of
Plant Physiology)
61 DUNKEL F LUNG P Z CHUAN L and YIN H F (1982) Insect and fungal responsc to sorbic
acid-treated wheat during storage in South China J €con Entomol 75(6) 1083-8
62. F E N h , G S (Nov 3, 1982) Apparatus and material for drying cut plants Eur Pat Appl Patrnt N o
63920
63. KERTESZETI EGYETEM (BUDAPEST). (Aug. 19, 1982). Titanium-containing phytoregulators (Japanese)
Jpn. Kokai Tokkyo Koho Patent N o 82134405.
64. CHUGAI PHARMACEUTICAL CO , LTD (Aug. 14, 1982). Insect disrase control with sorbit acid Jpn
Tokkyo Koho Patent No. 82 38222.
65 ROSEN, W.E., and BERKE, P.A. (Sept. '16, 1973). Modern concepts of c o s m ~ t i cprrsrrvdtion. J. Soc.
Cosmet. Chem. 24, 663-75.
66. EKLUND, T. (June 1983). The antimicrobial effect of dissociatPd and undissociatrd sorbic acid at
different p H levels. J. Appl. Bacteriol. 54(3), 383-9
67. BELL, T.A., ETCHELLS, J.L., and BORG, A.F. (1959). Influeiice of sorbic dcid on the growth of ( C r tain
species of bacteria, yeasts, and filamentous fungi. J Bacteriol 77, 573-80
68. YORK, G.K. (1960). Inhibition of microbes by sorbic acid. Ph.D. Thesis, University of California
69. SOFOS, J.N., and BUSTA, F F (1981). Antimicrobial activit) of sorbate. j. Food Prot 44(8), 611-22,647
70. DAHLE, H.K., and NORDAL, J. (1972). Effects of benzoic acid and sorbic acid on the production and
activities of some bacterial proteinases. Acta Agric. Scand. 22(1), 29-33.
71 TROLLER, J.A. (1965). Catalase inhibition as a possible mechanism of the fungistatic- action of sorbic
acid. Can. J. Microbio!. 11, 611-7.
72. AZUKAS, J.J., COSTILOW, R.N., and SADOFF, H.L. (1961) Inhibition of alcoholic fermentation by
sorbic acid. I.Bacteriol. 81,189-94.
73 AZUKAS, 1.1. (1963) Diss. Abstr. 23, 3588.
74.ANDERSON, T.E. (1963) Diss. Abstr 24, 2220
75 REINHARD, L., and RADLER, F. (1981). Effect of sorbic acid on Sdc~hdfon7\eSceiewside. I. Effect on
growth and the aerobic and anaerobic metabolism of glucose. Z Lebensm. Unters. For~ch.172(4),
278-83
76. REINHARD, L., and RADLER, F. (.1981). Effect of sorbic acid on Saccharom)re5 c-erewsiar 11. Adsorp-
tion, desorption and metabolism of sorbic acid. Z. Lebensm. Unters. Forsch. 172(5), 382-8
77 HARADA, K., HIZUCHI, R., and UTSUMI, I. (1968). Studies on sorbic acid Part I V Inhibition of thc
respiration in yeast. Agric. Biol. Chem. 32, 940-6.
78. DEAK, T., and NOVAK, E.K. (1970). Effect of sorbic acid on the growth of yeasts on various
carbohydrates. Acta Microbiol. 17(3), 257-66.
79. FREESE, E., SHEU, C.W., and GALLIERS, E. (1973). Function of lipophilic acids as antimicrobial food
additives. Nature 241, 321-5.
80.SHEU, C.W., SALOMON, D., SIMMONS, J.L., SREEVALSAN, T., and FREESE, E (1975) Inhibitory cfiects
of lipophilic acids and related compounds on bacteria and mammalian cells. Antimirrob. Agents
Chemother 7(3), 349-63
81 WEDDERBURN, D.L. (1958). Preservation of toilet preparations containing nonionics. J SOC. Coimct.
Chem. 9, 210-28
82 GUSTAVSSON, G. (1961). Farm. Aikakauslehti 70, 79-80.
83. NOWAK, G. (Nov. 15, 1962). Cer Pat. N o 1,'139,610 Chem. Abstr 58, 5152e (1963). Parfum. Cosmrt.
Savons 7, 69-75 (1964).
84. PATEL, N.K. (1967). Interaction of some pharmaceuticals with macromolecules. 1 1 1 . Corrclatiori of
binding data with inhibitory concentrations of preservatives in the presence of cetomacrogol 1000 and
polysorbate 80. Can. J. Pharm. Sci. 2(3), 77-80.
85 BAZILINSKAYA, N.V., and TARASOV, S.I. (1973). Effpcts of vitamins on the toxicity of sorbic acid to
Vvrtici//iurn dah/iae (Russian). Mikol. Fitopato!. 7(3), 205-8.
86 PATEL, N.K., and NACABHUSHAN, N. (Aug. 1969). Experiments in physical pharmacy. 111. Adsorption
by a plastic and in vitro biologic activity. Am. J. Pharm. Educ. 33, 392-9.
876 COSMETIC INGREDIENT REVIEW

87. MARTH, E.H., CAPP, C.M., HASENZAHL, L., JACKSON, H.W., and HUSSONC, R.V. (1966). Degradation
of potassium sorbate by Penicillium species. J. Dairy Sci. 49, 1197-1205.
88. DAVIDSON, P.M., BREKKE, C.J.,and BRANEN, A.L. (1981). Antimicrobial activity of butylated hydroxy-
anisole, teriary butylhydroquinone, and potassium sorbate in combination. J. Food Sci. &(I), 314-6
89. SEWARD, R.A., DEIBEL, R.H., and LINDSAY, R.C. (Nov. 1982). Effects of potassium sorbate and other
antibotulinal agents on germination and outgrowth of Clostridiurn bofulinurn type E spores in
microcultures. Appl. Environ. Microbiol. 44(5), 1212-21.
90.BEUCHAT, L.R. (Feb. 1981). Combined effects of solutes and food preservatives on rates of inactivation
of and colony formation by heated spores and vegetative cells of molds. Appl. Environ. Microbiol.
41(2), 472-7.
91 ROBACH, M.C. (1980). Interaction of salt, potassium sorbate and temperature on the outgrowth of
Closfridium sporogenes PA 3679 spores in a pre-reduced medium. J. Food Sci. 45(3), 742-3.
92. GOODING, C.M., and MELNICK, D. (1955). Sorbic acid as a fungistatis agent for foods. I X . Physico-
chemical considerations i n using sorbic acid to protect foods. Food Res. 20, 639-48.
93. RAO, G.K., MALATHI, M.A., and VIJAYAGHAVAN. (1966). Preservation and packaging of Indian foods.
11. Storage studies on preserved chapaties. Food Technol. 20, 1070-3.
94. KRAUZE, S., and FITAK, B. (1971). Influence of preservatives on the biosynthesis of nucleic acids and
on the protein content of animal cells in tissue culture (German). Mitt. Geb. Lebensrnittelunters. Hyg.
62(4), 359-67.
95. KAWANISHI, T., OHNO, Y., SUNOUCHI, M., ONODA, K., TAKAHASHI, A,, KASUYA, Y., and OMORI,
Y . (1981). Studies on nitrosamine formation by the interaction between drugs and nitrite. II. Hepatotox-
icity by the simultaneous administration of several drugs and nitrite. J . Toxicol Sci 6(4), 271-86.
96. CHMIELNICKA, J. (1967). Cabbage peroxidase. VII, Effects of preservatives on peroxidase and oxidase
activity. Acta Pol. Pharm. 24(4), 419-26.
97. ALIMUKHAMEDOVA, 0.5, and MAVLANI, M.I. (1977). Effect of antiseptics on the ultrastructural
organization of yeast cells (Ukrainian). Mikrobiol. Zh. (Kiev) 39(5), 651.
98. BARTHELMESS, A,, HUBER, H., and FUERST, H. (1968). Cytogenetic studies on natural antimutagenic
protective substances for radiodiagnosis, radiotherapy. and occupational radiation exposure. Strahlen-
therapie 136(1), 116-30.
99. HIERHOLZER, J.C., and KABARA, J.J. (1982). In vitro effects of monolaurin compounds on mveloped
RNA and D N A viruses. J. Food Safety 4(1), 1-12.
100. SHAPIRO, M., and BELL, R.A. (1982). Enhanced effectiveness of lyrnantria dispar (Lipidoptera. Ly-
mantriidae) nucleopolyhedrosis virus formulated with boric acid. Ann. Entomol. Soc. Am. 75(3), 346-9
101. SAITO, K. (1982). Inhibitory effects of dextranase produced by Chaetomium graoie on the develop-
ment of dental plaque and caries occurrence in rats (Japanese). Koku Eisei Gakkai Lasshi. 32(2),
147-66.
102. UCHIDA, O., NAITO, K., YASUHARA, K., OHBA, S., SATO, C., SHIMONURI, S., KOBAYASHI, K., and
TOBE, M . (1985). Studies on the acute oral toxicity of dehydroacetic acid, sorbic acid, and their
combination compound i n rats (Japanese). Bull. Natl. Inst. Hyg Sci. (Tokyo) 103(0), 166-71
103. SPARFEL, L., LAFILLE, C., and LE RESTE, S. (1968). Toxicological assays of some sorbic acid derivatives
(French). C.R. Hebd. Seances Acad. Sci. [D] 266(10), 1080-1.
'104. VERRETT, M., SCOTT, W.F, REYNALDO, E.F, ALTERMAN, E.K., and THOMAS, C.A. (1980). Toxic-ity and
teratogenicity of food additive chemicals i n the developing chicken embryo. Toxicol. Appl. Pharmacol.
56, 265-73.
'105. CTFA. (Oct. 4, 1983). Submission of unpublished data by CTFA. Acute oral toxicity. CTFA No. 3-1-29.
Available for review: Director, Cosmetic Ingredient Review, 1110 Vermont Ave., N.W.. Suite 810,
Washington, D.C. 20005
106. CTFA. (Dec. 19, 1977). submission of unpublished data by CTFA. Acute oral, dermal and ocular testing
o f CJ-0003, Lot TS1754. CTFA No. 3-1-10. Available for review Director, Cosmetic Ingredient Review,
11 10 Vermont Ave., N.W., Suite 810, Washington, D.C. 20005.
107. CTFA. (March 17, 1980). Submission of unpublished data by CTFA. Acute oral, ocular and 14-day
dermal toxicity testing of bronzer BC-0002, Lot PB-382-25-A. CTFA No. 3-1-14. Available for rcvicw:
Director, Cosmetic Ingredient Review, 1110 Vermont Ave., N.W., Suite 810. Washington, D.C 20005.
108. CTFA. (March 17, 1980). Submission of unpublished data by CTFA. Acute oral, ocular and '14-day
dermal testing of moisturizer AI-0032, Lot PB-381-4-C. CTFA No. 3-1-17. Available for review: Director,
Cosmetic Ingredient Review, 1110 Vermont Ave., N.W., Suite 810, Washington, D.C. 20005
ASSESSMENT: SORBIC ACID AND POTASSIUM SORBATE a77

109, 'SENIOR, A.E., and SHERRATT, H.S.A. (1969). A comparison of the effects on blood glucos? and
ketone-body levels, and of the toxicities, of pent-4-enoic acid and four simple fatty acids J Pharm.
Pharmacol. 21, 85-92.
110. FOOD A N D AGRICULTURE ORGANIZATION (FAO). (1971) United Nations Report Series (53A), 121
11.1, SHTENBERG, A.J., and ICNAT'EV, A.D. (1970). Toxicological evaluation of some combinations of food
preservatives. Food Cosmet. Toxicol. 8(4), 369-80.
112. CTFA. (Aug. 14,1974) Submission of unpublished data by CTFA. Safety evaluation of 0.5percent sorbic
acid in OlC/62445 (6016). four week subacute dermal toxicity study i n rabbits. CTFA No. 3-1-25.
Available for review: Director, Cosmetic Ingredient Review, 11 10 Vermont Ave., N . W , Suite 810.
Washington, D.C. 20005.
113. CTFA (Dec. 1979). Submission of unpublished data by CTFA. Subchronic (3-month) dermal toxicity
study of (CJ-0003) in rabbits, study: 6-7417. CTFA No. 3-1-11. Available for review: Director, Cosmetic
Ingredient Review, 1110 Vermont Ave., N.W., Suite 810, Washington, D.C 20005.
114. SNELL, K.C. (1967). Renal disease of the rat. In: Cotchin, E., and Roe, F.J.C. (eds.) fdrhology of
l a b o r a t o r y Rats and Mice. Oxford: Blackwell Scientific Publications, p. 105.
115. GAUNT, I.F., BUTTERWORTH, K.R., HARDY, 1.. and GANCOLLI, S.D. (1975). Long-term toxicitv of
sorbic acid in the rat. Food Cosmet. Toxicol. 13(1), 31-45
116. WENDY, R . J , HARDY, I., GAUNT, I.F., KISS, 1.5, and BUTTERWORTH, K.R. (1976). Long-term toxicity
studies of sorbic acid i n mice. Food Cosmet Toxicol. 14(5), 38'1-6.
117 BARCHENKO, I.P., and VASILIU, S.C. (1970). Effect of glucose oxidase. sorbic acid, and sodium
glutamate as food additives on the animal organism (Russian). Vop. Ratsion Pitan. 6 , 21-5.
118. O H N O , Y., SEKICAWA, S., YAMAMOTO, H., NAKAMORI, K . , and TSUBURA, Y. (1978). Additive
toxicity test of sorbic acid and benzoic acid i n rats (Japane~?). 1. Nara. Med. Assoc. 29(6), 695-708.
119. MASON, P.L., GAUNT, I.F., KISS, 1.5, HARDY, J., BUTTERWORTH, K.R., and GANCOLLI, S.D. (1976).
Long-term toxicity of parasorbic acid in rats. Food Cosmet. Toxicol 14(5), 387-94.
120. MASON, P.L., GAUNT, I.F., HARDY, J., KISS, 1.5, and BUTTERWORTH, K . R ('1976) Long-term toxicity
of parasorbic acid in mice. Food Cosmet. Toxicol. 14(5), 395-9.
121. BOURRINET, P., and RODDE, D. (1975). Determination of local tolerance to some preservative and
bactericidal agents (French) (from French translation). Labopharm. Prob. Tech 23(230), 121-4.
122. CTFA. (July 14, 1980). Submission of unpublished data by CTFA Eye irritation. CTFA No. 3-1-23,
Available for review: Director. Cosmetic Ingredient Review, 1 I10 Vermont Ave., N.W., Suite 810,
Washington, D C. 20005.
123 CTFA. (July 22, 1982). Submission of unpublished data by CTFA. Primary eye irritation of potassium
sorbate solution to rabbits. Study No. 820052. DMEH Project No. ML-82-083. CTFA No. 3-1-7. Available
for review: Director, Cosmetic Ingredient Review, 1110 Vermont Ave., N.W.. Suite 810, Washington,
D C. 20005.
124. CTFA. (Sept. 4, 1980). Submission of unpublished data by CTFA. CIR safety data test summary response
form: Sorbic acid, eye irritation. CTFA N o 3-1-3.
125. CTFA. (July 14, 1980). Submission of unpublished data by CTFA. Primary skin irritation. CTFA No.
3-1-22 Available for review: Director, Cosmetic Ingredient Review. I 1 10 Vermont AVP , N W., Suite
810, Washington, D.C. 20005.
126. CTFA. (June 9, 1975). Submission of unpublished data by CTFA. Primary skin irritation. CTFA N o
3-1-24. Available for review. Director, Cosmetic Ingredient Review, 1 I10 Vermont Ave., N.W., Suite
810, Washington, D.C. 20005.
127. MAURER, T., THOMANN, P., WEIRICH, E.C., and HESS, R. (1979) Predictive evaluation in animals of
the contact allergenic potential of medically important 5ubstances II. Comparlson of diffcrcnt
methods of cutaneous sensitization with "weak" allergens Contact Dermatitis 5(1), I - 10
128. NAMIKI, M., and KADA, T. (1975). Formation of ethylnitrolic acid by the reaction of sorbic acid with
sodium nitrite. Agric Biol Chem. 39, 1335
129. KADA, T. (1974). DNA damaging products from reaction between sodium nitrite and sorbic acid
Annu. Rep. Natl. lnst Genet. (24). 43-4.
130. HAYATSU, H., CHUNG, K.C., KADA, T., and NAKAJIMA, T. (1975). Ccnrration of mutagenic tom-
pound(s) by a reaction between sorbic acid and nitrite. Mutat. R r s 30(3), 317-9.
I31 OSAWA, T., and NAMIKI, M. (1982). Mutagen formation in the reaction of nitrite with the food
components analogous to sorbic acid. Agric Biol. Chem. 46(9), 2299-304.
132 PIERSON, M.D , and SMOOT, L.A. (1982). Nitrite, nitrite alternatives, and the control of Clostridium
botulinum in cured meats. CRC Crit. Rev. Food Sci Nutr. 17(2), 111-87.
878 COSMETIC INGREDIENT REVIEW

133. DIFATE, V.C. (1977). Microbiological Mutagenicity Evaluation of Sorbic Acid/Sodium Nitrite Reaction
Products, report submitted to the Expert Panel on Nitrites, Nitrates, and Nitrosamines, U.S. Department
o f Agriculture. Washington, D C.
134. ROBACH, M.C., DIFATE, V.G., ADAM, K., and KIER, L.D. (1980). Evaluation of the mutagenicity of
sorbic acid-sodium nitrite reaction products produced i n bacon-curing brines. Food Cosmet. Toxicol
18(3), 237-40
135. KHOUDOKORMOFF, B., and GIST-BROCADES, N.V. (1978). Potential carcinogenicity of some food
preservatives in the presence of traces of nitrite. Mutat. Res. 53(2), 208-9.
136. NAMIKI, M., UDAKA, S., OSAWA, T., TSUJI, K., and KADA, T. (1980). Formation of mutagens by sorbic
acid-nitrite reaction: Effects of reaction conditions on biological activities Mutat. Res. 73(1), 21-8.
137. NAMIKI, M., OSAWA, T., ISHIBASHI, H., NAMIKI, K., and TSUjI, K. (1981). Chemical aspects of
mutagen formation by sorbic acid-sodium nitrite reaction. J. Agric Food Chern. 29(2), 407-11.
'138. TANAKA, K., CHUNG, K.C., HAYATSU, H., and KADA, T. (1978). Inhibition of nitrosamine formation in
vitro by sorbic acid. Food Cosmet. Toxicol. 16(3), 209-16.
139. LATHIA, D., and SCHELLHOEH, B. (1981). Inhibition of nitrosamine formation i n vitro i n prewnce of
both ascorbic acid and sorbic acid. Recent Adv. Clin. Nutr. 1, 189-90.
-140. MASSEY, R.C., FORSYTHE, L., and MCWEENY, D.J. (1982). The effects of ascorbic acid and sorbic acid
on N-nitrosamine formation in a heterogeneous model system. J. Sci. Food Agric. 33(3), 294-8.
141. AMUNDSON. C.M., SEBRANKEK, J.C., RUST, R.E., KRAFT, A.A., WAGNER, M.K., and ROBACH, M.C.
(1981). Effect of belly composition on sorbate-cured bacon. J. Food Sci. 47(1), 218-21
142. KAWANISHI, T., OHNO, Y., TAKAHASHI, A,, ISHIWATA, H., TANIMURA, A,, KASUYA, Y., and
OMORI, Y. (1981). Studies on nitrosamine formation by the interaction between drugs and nitrite. I.
Measurement of the amount of nitrosamine formed in rat and guinea pig stomachs. J. Toxicol Sci. 6(4),
261 -86.
143 RAO, C.S., OSBORN, J.C., and ADATIA, M.R. (1982). Drug-nitrite interactions in human saliva: Effects
of food constituents on carcinogenic N-nitrosamine formation. J. Dent. Res. 61(6), 768-71
144 KABACOFF, B.L., DOUGLASS, M.L., ROSENBERG, I.E., LEVAN, L.W., PUNWAR, J.K , VIELHUBER, S F ,
and LECHNER. R.J. (1984). Formation of nitrosamines in non-ionic and anionic emulsions in the
presence and absence of inhibitors. IARC Sci. Publ. 57, 347-52.
145. D'AQUINO, M., and SANTINI, P. (1977). Food additives and their possible genetic toxicity; microbio-
logical determination (Spanish). Arch. Latinoam. Nutr. 27(3), 411-24.
146. MORITA, K., ISHIGAKI, M., and ABE, T (1981). Mutagenicity of materials related with cosmetics. J. Soc
Cosmet. Chem. japan 15(3), 243-53.
147. KAWACHI, T., YAHACI, T., KADA, T., TAZIMA, Y., ISHIDATE, M., SASAKI, M., and SUCIYAMA, 5
(1980). Cooperative program on short-term assays for carcinogenicity in japan. IARC (Int. Agency Res
Cancer) Sci. Publ. 27, 323-30.
148. ABE, S., and SASAKI, M. (1977). Chromosome aberrations and sister chromatid exchanges in Chinese
hamster cells exposed to various chambers. JNCl 58(6), 1635-4'1.
149. ISHIDATE J R , M., SOFUNI, T., YOSHIKAWA, K., HAYASHI, M., NOHMI, T., SAWADA, M , and
MATSUOKA, A . (1984). Primary mutagenicity screening of food additives currently used in Japan. Food
Chem. Toxicol. 22(8), 623-36
150. LITTON BIONETICS. (Nov. 25, 1974). Mutagenic Evaluation of Compound FDA 73-4, Potassium
Sorbate. Prepared for the Food and Drug Administration, Contract No. 223-74-2104. NTlS Document
No. PB-245 434. Kensington, M D .
151. HASEGAWA, M.M., NISHI, Y , OHKAWA, Y . , and INUI, N. (1984). Effects of sorbir acid and its salt5 on
chromosome aberrations, sister chromatid exchanges and gene mutations i n cultured Chinese hamster
cells. Food Chem. Toxicol. 22(7), 501-7.
152. TSUCHIYA, T., and YAMAHA, T. (1983). The production of mutagens in the intestine of mice fed on a
diet containing 15 percent sorbic acid (I). J. Toxicol. Sci. 8(1), 15-24.
153. TSUCHIYA, T., and YAMAHA, T. (1983). Urinary excretion of mutagens and the effects of sorbic acid on
the lipid peroxide level in the mice fed on a 15 percent sorbic acid diet. J. Toxicol. Sci. 8(3), 213-22.
154. TSUCHIYA, T., and YAMAHA, T. (1984). Depletion of the reduced glutathione level in the liver and
production of the mutagens in the intestine in the mice inducing hepatoma by feeding on a hlgh level
dose of sorbic acid. Mutat. Res. 130(4), 267-72.
155. ISHIZAWA, K . , SHIBUYA, K., GON, S., and AISO, K. (1980). Poisonous effect of the sorbic acid on the
mouse fed i n long-term with its high concentration (secondary reference). J. Tokyo Kasei Cakuin
College 20, 83-97.
ASSESSMENT: SORBIC ACID AND POTASSIUM SORBATE 879

156.FDRL. (1975). Teratologic evaluation of FDA 73-3 (potassium sorbate: Sorbistat) in mice and rats.
Prepared under DHEW Contract No. FDA 223-74-2176 NTlS No PB-245520 Waverly, NY, 32 pp.
157. CLEMMENSCN, O., and HJORTH, N. (1982). Perioral contact urticaria from sorbic acid and benzoic ac-id
i n a salad dressing Contact Dermatitis 8(l), 1-6
158. SOSCHIN, D., and LEYDEN, 1.1. (1986). Sorbic acid-induced erythema and cdema J Am Acad.
Dermatol. 14, 234-41
159. MARZULLI, F.N., and MAIBACH, H.I. (1973). Antimicrobials. Experimental contact sensitization in man
J. SOC. Cosmet Chem. 24(7), 399-421.
160. MARZULLI, F.N., and MAIBACH, H.I. (1974). Use of graded concentration in studying skin sensitizers.
Experimental contact sensitization in man. food Cosrnet. Toxicol. 1 2 ( L ) , 219-27
I 6 1 KLAUDER, J.V. (1960). The closed, open, prophetic and repeated insult patch tests in studv of
cutaneous reaction. J Soc. Cosmet. Chem. 11, 249-68.
162. CTFA. (June 2, 1981). Submission of unpublished data by CTFA. CIR safety data tcst summary response
form: Sorbic acid in product, skin irritation. CTFA No. 3-1-20.
‘163. CTFA. (Feb. 23,1981). Submission of unpublishpd data by CTFA. CIR safpty data test summar\’ response
form. Sorbic acid in product, skin irritation. CTFA No. 3-1-21
164. CTFA. (Feb. 12, 1982) Submission of unpublished data by CTFA. Allcrgic contact seiisitization test (Tc5t
No 191-82) 0.5 percent Sorbic Acid in OlC/62445-09 (6016). CTFA No. 3-1-26 Available for review
Director, Cosmetic Ingredient Review, 1110 Vermont Ave, N.W., Suite 810, Washington, D C. 20005
165. CTFA. (April 6, 1979). Submission of unpublished data by CTFA. Alltrgic contact sensitization tc’st ( T w t
No 075-79) 02B/15146-02 containing 0 2 percent sorbic acid CTFA No. 3-1-27. Available for rcwc’w:
Director, Cosmetic Ingredient Review, 1110 Vermont Ave., N.W., Suite 810, Washington, D C. 20005.
166. CTFA. (June 1, 1979) Submission of unpublished data by CTFA. Allergic contact sensitization test (Test
No 079-79) containing 0.2 percent sorbic acid in 02D-15148-06 CTFA N o 3-1-28. Available for review:
Director, Cosmetic Ingredient Review, 1110 Vermont Ave., N.W., Suite 810, Washington, D C 20005
167. CTFA. (Feb. 18, 1983). Submission of unpublished data by CTFA. Allergic contact sensitization test (TPst
No. 237-83) 0.2 percent sorbic acid in 12F/21526-21. CTFA No. 3-1-30. Available for reviPw Dirrctor,
Cosmetic Ingredient Review, 1110 Vermont Ave., N.W., Suite 810, Washington, D.C. 20005
168. CTFA (Oct. 22, 1982). Submission of unpublished data bv CTFA Allcrgic contact sensitization tcst
(-rest No. 193-82) 0.2 percent sorbic acid in 12J/21526-19 CTFA No. 3-1-31. Available for re\ie\v
Director, Cosmetic Ingredient Review, 1110 Vermont Ave., N W., Suite 810, Washington, D C 20005
169. C.TFA. (Dec. 5, 1983) Submission of unpublished data bv CTFA Report of a human skin test of
cumulative irritation. CTFA No. 3-1-12. Available for review Director, Cosmetlc Ingredient Review,
1110 Vermont Ave., N.W., Suite 810, Washington, D C. 20005
170 CTFA (Dec. 13, 1979). Submission of unpublished data bv CTFA. The study of cumulative irritant
properties of a series of test materials. CTFA N o 3-1-15. Available for review Director, Cosmetic
Ingredient Review, 1110 Vermont Ave., N.W., Suite 810, Wa3hington. D C. 20005.
171. CTFA. (Dec. 13, 1979). Submission of unpublished data by CTFA. The study of cumulative irritant
properties of a series of test materials. CTFA N o 3-1-18 Available for review. Director, Cosmetic
Ingredient Review, 11.10 Vermont Ave., N.W., Suite 810, Washington, D C. 20005.
172. CTFA. (May 13, 1986). Submission of unpublished data bv CTFA. Protocol, repeated insult patc-h
testing. CTFA No. 3-1-13. Available for review: Director, Cosmetic Ingredient Review, 1 I10 Vermont
Ave., N.W., Suite 810, Washington, D.C. 20005.
173. CTFA. (May 13,1986). Submission of unpublished data by CTFA Protocol, repeated insult patch testing
of bronzer. CTFA No. 3-1-16. Available for review: Director, Cosmetic Ingredient R e ~ i e w ,1 1 10
Lermont Ave., N.W., Suite 810, Washington, D C. 20005
174. CTFA. (May 13, 1986) Submission of unpublished data by CTFA Protocol, repeated insult patch testing
of moisturizer CTFA N o 3-1-19, Available for review: Direc-tor, Cosmetic Ingredient Review, I I 10
Vermont Ave., N.W., Suite 810, Washington, D.C. 20005.
175. C:Tf:A. (Aug. 24, 1983). Submission of unpublished data by CTFA Report N o PI-3102, ELaluation of
potential hazards by dermal contact. Test material: SCS-10, CMEH No 2173 CTFA No. 3-1-31 Available
for review: Director, Cosmetic Ingredient Review, 1110 Vermont Avc, N W., Suite 810, Washington,
D.C. 20005.
176. CTFA. (Aug 24, 1983). Submission of unpublished data by CTFA. Report No PI-3101, Evaluation of
potential hazards by dermal contact, Test material: SCS-’13, CMEH No. 2173 CTFA N o 3-1-33. Available
for review: Director, Cosmetic Ingredient Review, 1110 Vermont Ave , N.W., Suite 810, Washington,
D.C. 20005.
880 COSMETIC INGREDIENT REVIEW

177. CTFA. (July 25, 1983). Submission of unpublished data by CTFA. CMEH 2160, SCS-5002-0 U.K.,
Repeated insult patch test, 56 subjects, Final Report, H R L 83-114. CTFA No. 3-1-34. Available for
review: Director, Cosmetic Ingredient Review, 1110 Vermont Ave., N.W., Suite 810, Washington, D.C.
20005,
178. CTFA. (July 20, 1983). Submission of unpublished data by CTFA. Report GMEH 2157-8, Repeated insult
patch test for evaluation of dermal irritancy/sensitization in human subjects CTFA No. 3-1-35.
Available for review: Director, Cosmetic Ingredient Review, 1110 Vermont Ave , N.W., Suite 810,
Washington, D.C. 20005.