EFFECT OF BLANCHING TIME ON SELECTIVE MINERAL ELEMENTS EXTRACTION FROM THE SPINACH SUBSTITUTE COMMONLY USED IN BRAZIL

EFFECT OF BLANCHING TIME ON SELECTIVE MINERAL ELEMENTS
EXTRACTION FROM THE SPINACH SUBSTITUTE (Tetragonia expansa)
COMMONLY USED IN BRAZIL1
Luciane M. KAWASHIMA2 , Lucia M.VALENTE SOARES2 ,*
SUMMARY
The true spinach (Spinacia olerac

ea) does not grow well in warm climates and for that reason is not commercialized in Brazil.
Instead, a spinach substitute (Tetragonia expansa ,
) originally from New Zealand, is widely used in the country. There is scant
information on the mineral profile and none on the soluble mineral fraction of this vegetable. The solubility of a mineral is one of
the important factors for its absorption. For this reason, the calcium, magnesium, iron, manganese, copper, zinc, potassium, and
sodium soluble fractions in the raw spinach substitute were determined and the effect of blanching times on the solubility of these
minerals was investigated. Blanching times of 1, 5, and 15 minutes were employed. The magnesium, manganese, potassium, and
sodium soluble fractions increased sizably with shorter blanching time. Longer blanching time (15 minutes) caused large losses of
minerals. The soluble mineral fractions can contribute poorly to diet in terms of potassium, magnesium, manganese, and zinc.
The spinach substitute cannot be considered a dietary source of calcium, iron and copper due to the insolubility of these minerals
in the vegetable, possibly caused by the large oxalate content.
Keywords: spinach substitute; effect of blanching; mineral elements.
RESUMO
EFEITO DO TEMPO DE BRANQUEAMENTO NA EXTRAÇÃO SELETIVA DE ELEMENTOS MINERAIS DO SUBSTITUTO DE
ESPINAFRE (Tetragonia expans C
a) OMUMENTE EMPREGADO NO BRASIL. O verdadeiro espinafre (Spinacia oleracea )não se de-
senvolve bem em climas quentes e por esta razão não é comercializado no Brasil. Em seu lugar, um substituto de espinafre (Tetra-
gonia expansa), originário da Nova Zelândia, é amplamente utilizado. A informação sobre o perfil mineral é resumida e inexistente
sobre a fração solúvel de minerais deste vegetal. A solubilidade de um mineral é um dos fatores importantes para sua absorção.
Por esta razão, as frações solúveis de cálcio, magnésio, ferro, manganês, cobre, zinco, potássio e sódio foram determinadas no
substituto de espinafre cru, e o efeito dos tempos de branqueamento na solubilidade destes minerais foi investigado. Tempos de
branqueamento de 1, 5, e 15 minutos foram empregados. As frações solúveis de magnésio, manganês, potássio, e sódio, aumenta-
ram consideravelmente com o menor tempo de branqueamento. Um período mais prolongado de branqueamento (15 minutos) cau-
sou perdas sensíveis de minerais. As frações solúveis dos minerais podem ter pequena contribuição para a dieta em termos de po-
tássio, magnésio, manganês, e zinco. O substituto de espinafre não pode ser considerado como uma fonte de cálcio, ferro e cobre
para a dieta devido à insolubilidade destes minerais no vegetal, possivelmente causada pelo elevado teor de oxalato.
Palavras-chave: substituto de espinafre; efeito de branqueamento; elementos minerais.
1 - INTRODUCTION
dependent upon their chemical form and the presence of
enhancers or inhibitors within the food as well as the
In Brazil, a spinach substitute (Tetragonia expansa)
nutritional status and requirements of the individual
is used for salads and cooking. It originated in New
ingesting the food [3,4,6]. In vitr
o and in vivo tests have
Zealand and is a succulent belonging to the Aizoaceae
been used to measure mineral availability. In vitr
o
family. It grows well in tropical climates and for that
methods involve a simulation of the digestive process
reason it well adapted in this country. The true spinach
and the information supplied is partial because it does
(Spinacia olerac
, o
ea) n the other hand, a leafy vegetable
not include the utilization of the mineral by the human
of the Chenopodiaceae family, is commonly consumed in
organism. In viv s
o tudies measure either radioactive or
the temperate regions of the globe [5] and does not grow
stable isotopes in human subjects or animals [1,3,12].
well in warm climates. The true spinach is not They are closer to the natural absorption process
commercialized in Brazil. The concentrations of a few
although lengthier and more expensive to be conducted
nutritionally important minerals and their absorbability
than in vitro methods. Estimates of availability through
have been studied for the true spinach (Spinacia
mathematical models that take into account the
oleracea [
) 2,8,14]. Recently, the mineral composition for
concentrations of minerals and those of absorption
the New Zealand spinach substitute used in Brazil and of
inhibitors and enhancers have also been proposed [20].
other leafy vegetables has been published [9].
Solubility of the mineral is one of the conditions for
The presence of a nutrient in the food does not mean
absorption by the digestive tract and its knowledge
availability. The absorption of mineral nutrients is provides initial information about its potentiality for
absorption [2,10]. Solubility measurements are simpler
1 Recebido para publicação em 11/09/2002. Aceito para publicação
and less costly to be conducted than in viv an
o
d in vitr
o
em 06/04/2005 (000993).
studies. In this line TESSIER, CAMPBELL & BISSON [18]
2 Department of Food Science, State University of Campinas, C.P.
proposed an interesting protocol first used for the study
6121, CEP: 13081-970 - Campinas, SP, Brazil.
of river sediments. It consists of a simplified method by
* A quem a correspondência deve ser enviada.
extracting selectively minerals under soluble form,
Ciênc. Tecnol. Aliment., Campinas, 25(3): 419-424, jul.-set. 2005
419

---

Effect of blanching on spinach mineral elements, Kawashima & Valente Soares
bound to ligands with pK 5,
? bound to ligands with pK ?
homogenized with the help of a food processor. The food
2, bound to organic insoluble ligands, and residual. This
processor parts in contact with samples were treated
approach has been applied to experimental animal diets
with a hot aqueous solution of 2% EDTA and 2% citric
and gastrointestinal contents in the study of iron acid [13] in order to remove any trace of metals, washed
digestion and absorption in rats [17]. REYKDAL & LEE
with de-ionized water and air dried before use.
[14] followed another approach and measured dialyzable
After washing and drying and before comminuting,
calcium in milk and true spinach in samples submitted
250g portions of the samples, leaves and tender stems,
to HCl and pepsin in vitr
o digestion and non-digested
were separated for blanching. The leaves and stems were
samples. The dialyzable calcium content was more than
heated in a glass pan containing 2L of boiling de-ionized
twice as large in digested samples when compared with
water for periods of 1, 5, and 15 minutes, respectively.
non-digested samples. SCHMIDT, MACDONALD & The blanched vegetable was then dried in a hand
KELLY [16], by their turn, measured iron, calcium, and
spinning kitchen vegetable drier and homogenized in a
magnesium in water extracts and in extracts submitted
food processor as described above.
to in vitr p
o eptic digestion obtained from raw and cooked
amaranth and collard leaves. The results from both
2.2 - Total solids
procedures were close for the raw samples, but in the
case of the cooked samples the results were twice as large
Duplicate determinations were conducted in raw
in the peptic digested extracts when compared with the
and blanched samples by drying the ground sample in an
o
water extracts.
aluminum dish in an atmospheric oven at 105 C until
constant weight. Fifteen mg/cm2 of diatomaceous earth
Oxalates are an important inhibitor to mineral were added to each aluminum dish for better heat
absorption because they form insoluble salts with distribution. The total solids determination was used to
calcium, iron, and magnesium. The presence of oxalates
calculate the minerals concentration on dry basis and to
in foods is of concern, especially for calcium, as they
compare the levels of elements among the sequentially
reduce the absorbability of this mineral [19]. Calcium
extracted fractions.
absorbability from true spinach, a high oxalate food, has
been shown to be low for human subjects [8]. The
2.3 - Determination of individual minerals
amount of oxalate present in the raw New Zealand
spinach substitute, also used in Brazil, has been
Total contents of each mineral were determined in
determined and it was approximately 5 times higher
duplicates after digestion of the samples with nitric acid.
(1764.7 ± 95.7mg/100g) than that of the true spinach
After complete destruction of organic matter, the residue
(329.6 ± 0.8mg/100g) [15]. Note that the standard was dissolved with enough nitric acid and lanthanum
deviation indicates the variability encountered in nature
solution in order to reach a final concentration of 1%
and not the uncertainty of the determination.
HNO3 + 0.5% lanthanum in a volumetric flask of
appropriate size. The solution was used to determine
The present research aimed at (a) fractionating six
potassium, sodium, calcium, magnesium, iron, copper,
nutritionally important minerals (Ca, Mg, Mn, Fe, Zn and
manganese, and zinc by flame atomic absorption
Cu) and two electrolytes (K and Na) in the spinach
spectrometry on a microprocessor controlled Perkin-
substitute used in Brazil in order to verify the Elmer instrument, model 5100 PC with deuterium
distribution profile of the minerals in the soluble fraction
background correction lamp. The instrument operating
(extractable at pH 7.0), in the fraction bound to insoluble
conditions used in determining each element can be
ligands (capable of solubilization by a mild peroxide and
found in Table . An
1
air acetylene flame was used and
acid treatment), and in the residual fraction (resistant to
standards for calibration were from Carlo Erba, Merck
mild peroxide and acid treatment) and (b) the effect of
and J.T. Baker. The standards were prepared with the
increasing blanching times on the mineral profile of
same concentration of nitric acid and lanthanum as the
these fractions and on possible losses by leaching.
samples. De-ionized water (Milli-Q) with 18 megaohm of
2 - MATERIAL AND METHODS
resistivity was used whenever water was need.
Glassware and polyethylene flasks were soaked 24 hours
2.1 - Sample preparation
in nitric acid 10% and then washed with de-ionized water
before use. The individual concentrations of each
Three bunches of spinach substitute (Tetragonia
mineral in each sample were calculated on wet and dry
expansia
) were acquired in markets in the city of weight basis. The data on dry basis was used to calculate
Campinas, state of São Paulo, five times during the
the concentration of each element in each extraction
period of six months. The bunches of leaves were washed
fraction as well as losses to blanching water for each
under running water to remove sandy particles and then
individual sample.
washed with de-ionized water, dried in a hand spinning
kitchen vegetable drier, and patted dry with paper 2.4 - Sequential extraction of minerals
towels. Hard stems and wilted leaves were removed. The
The soluble fractions of individual minerals and
leaves and tender stems were comminuted, well mixed by
individual minerals bound to insoluble ligands were
hand and about 250g were removed and further
420
Ciênc. Tecnol. Aliment., Campinas, 25(3): 419-424, jul.-set. 2005

---

Effect of blanching on spinach mineral elements, Kawashima & Valente Soares
TABLE 1 - Operating conditions used for determining each
2.5 - Analytical quality control
element
The detection limits for each element were
determined preparing two solutions of the element, one 5
times and the other 2 times the expected detection limit.
Twenty readings were made of each solution in
alternation with a blank. The detection limit was
calculated multiplying the concentration of the standard
by 3 times the standard deviation obtained. The result
was divided by the mean of the determinations.
determined [18]. The soluble fraction was defined as the
relationship between the concentrations of the element
A blank and an internal reference sample were
extractable at pH 7.0 and its total concentration in the
added to each batch of samples analyzed as an every day
spinach substitute. For the soluble fraction of minerals,
quality control check. The certified reference material
three replicates were prepared for each sample. The
1573a Tomato Leaves (NIST, Gaithersburg, U.S.A.) was
sample was weighed (10g) into a polypropylene used to check the accuracy of the instrument and of the
centrifuge tube and shaken one hour with 8mL 0.15M
procedure employed to determine the total values of the
ammonium acetate solution (pH 7,0). After centrifuging
elements being studied.
30 minutes at 10,000rpm the supernatant was
3 - RESULTS AND DISCUSSION
separated and the residue in the tube washed with 10mL
de-ionized water. The tube with water and residue was
The limits of detection and the analytical
again centrifuged at 10,000rpm for 15 minutes. The
uncertainties under the experimental conditions are
second supernatant was collected and added to the first
presented in Table .
2 The average standard deviation for
and this was labeled soluble fraction. Each mineral was
duplicates of the same sample expresses the analytical
determined in the soluble fraction by atomic absorption
uncertainties of the analyses and indicates the number
spectrometry as described for the total minerals content
of significant digits of each individual result.
determination.
TABLE 2 - Limits of detection and analytical uncertainty under
The fraction bound to insoluble ligands was defined
the conditions employed during the present work
as the relationship between the concentration of the
element extractable after mild peroxide and acid
treatment and its concentration in the spinach
substitute. The minerals bound to insoluble ligands were
determined in the residue from the previous extraction
by adding 3mL of 0.02M HNO3 and 5mL of 30% H2 O2
adjusted to pH 2 with HNO
Five samplings were conducted during six months
3 . The mixture was heated to
85o C for two hours in a water bath and shaken and the variation was expressed as the standard
occasionally. After that period, another 3 mL portion of
deviation among samples (Table 3 .) A great variability in
30% H
o
the concentration of the minerals was found among the
2 O2 was added. The mixture was kept at 85 C for
another period of 3 hours with occasional shakings. samples of the vegetable throughout the six months of
After that the mixture was allowed to cool and 5 mL of 3.2
study. This variability between samples is of one or two
M ammonium acetate in 20% HNO
orders of magnitude greater than the analytical
3 were added. The
mixture was diluted to 20mL with de-ionized water and
uncertainty (Table 2 ,
) except in cases the mineral present
shaken for 30 minutes. The contents of the flask were
in the fraction analyzed is at trace level (Fe, Mn, Cu and
allowed to settle and the supernatant was collected for
Zn). Such data gives support to the analytical results and
the minerals determination as described above.
to the conclusions to be derived. Interestingly, the total
solids content varied among samples from 4.82 to
The residual fraction was defined as the relationship
7.60g/100g, with an average and standard deviation of
between the concentration of the element not extractable
6.08 ± 1.14g/100g. The variation among samples was
after mild peroxide and acid treatment and its total
smaller than the one found also among samples for the
concentration in the spinach substitute. The residual
minerals present in higher content (Ca, Mg, K and Na).
fraction for each element was calculated subtracting the
values of soluble and insoluble fractions from the total
The concentration of soluble calcium was very low
concentration of the respective element.
(Table 3) in the spinach substitute. A lengthy blanching
time such as 15 minutes (Figure
d
1) id not improve the
Losses of a mineral to the blanching water by solubility of this mineral. The fraction bound to insoluble
leaching were calculated as the difference between the
ligands but capable of sulubilization by mild acidic and
total amount of the element found in the raw sample and
oxidizing conditions was also low (Figure . T
2) he residual
the total amount found in the blanched sample. The
fraction in the raw vegetable contained 98% of the total
results were calculated for each individual sampling and
content of calcium (Figure

3) and varied little with
averaged.
Ciênc. Tecnol. Aliment., Campinas, 25(3): 419-424, jul.-set. 2005
421

---

Effect of blanching on spinach mineral elements, Kawashima & Valente Soares
TABLE 3 - Individual element concentrations in the fractions of the raw spinach substitute (Tetragonia expansa)
blanching. In such situation it is expected that calcium
in the spinach substitute would not contribute to the
diet. The high insolubility of calcium in this vegetable
might be explained by the high content of insoluble
oxalate as reported for the New Zealand spinach [15].
FIGURE 3 - Average changes in the residual fraction of minerals
caused by increasing blanching time. The residual fraction
(%) was defined as the ratio between the concentration of the
element not extractable after mild peroxide and acid
treatment and its total concentration in the raw spinach
substitute
Magnesium was mostly in soluble form. The soluble
fraction of magnesium in the raw vegetable contained
20mg/100g (Table
a
2) nd represented about 37% of the
FIGURE 1 - Average changes in the soluble fraction of minerals
caused by increasing blanching time. The soluble fraction (%)
total content of the mineral (Figure 1). The RDA for
was defined as the ratio between the concentrations of the
magnesium for adults of both sexes is 4.5mg/kg of body
element extractable at pH 7.0 and its total concentration in
weight [11]. The data obtained on solubility indicated
the raw spinach substitute
that the contribution of magnesium by the spinach
substitute is small. Blanching for 1 or 5 minutes
increased the soluble amounts of the mineral (Figure 1 .)
But a noticeable decrease in the soluble fraction
occurred with blanching for 15 minutes due to losses of
over 50% to the water by leaching (Table .
4)
TABLE 4 - Average mineral losses (%) in the spinach substitute
(Tetragonia expansa) to the water by leaching
FIGURE 2 - Average changes in the fraction of minerals bound
For iron, another mineral with solubility affected by
to insoluble ligands caused by increasing blanching time. The
the presence of oxalates, a concentration of
fraction bound to insoluble ligands (%) was defined as the
0.06mg/100g was found in the soluble fraction of the
ratio between the concentration of the element extractable
raw edible parts (Table 2 .
)
The soluble fraction
after mild peroxide and acid treatment and its concentration
in the raw spinach substitute
represented 8% of the total iron concentration in the raw
422
Ciênc. Tecnol. Aliment., Campinas, 25(3): 419-424, jul.-set. 2005

---

Effect of blanching on spinach mineral elements, Kawashima & Valente Soares
vegetable and this fraction doubled with 1 minute vegetable was 28mg/100g and represented 44% of the
blanching. The soluble fraction did not exhibit further
total sodium present (Table
. T
3)
he concentrations of
increase after 5 or 15 minutes blanching (Figure
.
1)
soluble sodium increased with blanching times (Figure
Losses due to leaching were low (Table
w
4)
ith any
1). Longer blanching times increased losses of sodium to
blanching time.
the blanching water and these losses ranged from 26 to
The soluble fraction of manganese (Table 3
)
57% (Table 4 .) Longer blanching times reduced the
contained 0.3mg/100g. The recommended daily intake
residual fraction and the fraction bound to insoluble
for manganese for adults ranged from 2.0 to 5.0mg [11],
ligands was smaller in the blanched vegetable when
independently of body weight, what pointed to a rather
compared with the same fraction in the raw spinach
low possible contribution to diet by this spinach substitute, suggesting both fractions served as the
substitute. The soluble fraction of manganese source of sodium for the soluble fraction (Figures 2 and
3)
represented 34% of this mineral in the raw vegetable. The
as leaching increased due to longer blanching.
data indicated that blanching caused losses of about
4 - CONCLUSIONS
14% when the vegetable was submitted up to 5 minutes
blanching and higher (42%) for 15 minutes blanching
For the New Zealand spinach substitute commonly
from the total amount of manganese initially present in
used in Brazil, a shorter blanching time of one minute is
the raw vegetable (Table 4).
the most beneficial in yielding higher concentrations of
soluble minerals. However, the vegetable is a poor source
The levels of copper (0.02mg/100g) in the soluble
for soluble potassium, magnesium, and manganese; an
fraction (45%) of the raw vegetable could only contribute
even meager source for zinc and cannot be considered a
poorly to diet (Table 3). The recommended daily intake for
dietary source for calcium, iron, and copper.
copper ranges from 1.5 to 3mg daily for adults of any
body weight [11] and consequently the spinach
5 - REFERENCES
substitute does not qualify as a source for copper in the
diet. The soluble fraction increased with blanching at the
[1]
BARBERÁ, R.; FARRÉ, R. Biodisponibilidad de los
expense of the residual and organic bound fractions
elementos traza. Revista Española de Ciencia y
(Figures 1, 2 and 3). Calculated losses to the blanching
Tecnología de Alimentos, v. 34, n. 4, p. 381-399, 1992.
water were small and did not increase much with [2]
BENWAY, D.A.; WEAVER, C.M. Assessing chemical form
blanching time (Table 4 .
)
of calcium in wheat, spinach, and kale. Journal of Food
Science, v. 50, n. 3, p. 605-608, 1993.
The concentration of solubIe zinc in the raw
vegetable was 0.12mg/100g (Table 3), too low to make
[3]
COZZOLINO, S.M.F. Biodisponibilidade de minerais.
any sensible contribution to diet. Zinc deficiency has
Revista de Nutrição da PUCCAM ,
P v. 10, n. 2, p. 87-98,
1997.
been a matter of concern for developing countries [7]. The
recommended daily intakes for zinc are 12 and 15mg for
[4]
EKMEKCIOGLU, C. Intestinal bioavailability of minerals
women and men, respectively, independent of body
and trace elements from milk and beverage in humans.
weight [11]. Zinc was present in the soluble fraction of
Nahrung, v. 44, n. 6, p. 390S-397S, 2000.
the raw vegetable at about 34% of the total concentration
[5]
ENCYCLOPAEDIA BRITANNICA, INC. Encyclopaedia
t h
of the element. Blanching decreased the amount of the
Britannica. 15 edition, Chicago, v. 3, p. 375, 1979.
element in the soluble fraction (Figure 1 .
) The losses of
[6]
FAIRWEATHER-TAIT, S.J. Bioavailability of dietary
zinc to the blanching water increased with longer
minerals. Biochemical Society Transactions, v. 24,
blanching times (Table 4 .
)
p. 775-780, 1996.
The raw vegetable contained 200 mg of soluble
[7]
GIBSON, R.S. Zinc nutrition in developing countries.
potassium /100g (Table 3). The RDA for potassium is
Nutrition Research Reviews, v. 7, p. 151-173, 1994.
3500mg for adults of any body weight [11] and so the
[8]
HEANEY, R.P.; WEAVER, C.M.; RECKER, R.R. Calcium
contribution of this element to diet by this spinach
absorbability from spinach. American Journal of
substitute is marginal. Surprisingly the soluble fraction
Clinical Nutritio , v
n . 47, p. 707-709, 1988.
contains 39% of the total amount of the mineral [9]
KAWASHIMA, L.M.; VALENTE SOARES, L.M. Mineral
indicating that part of the mineral is connected to
profile of raw and cooked leafy vegetables consumed in
insoluble structures and is not readily accessible in the
Southern Brazil. Journal of Food Composition and
raw vegetable. Blanching for 1 minute and 5 minutes
Analysis, v. 16, p. 605-611, 2003.
increased the concentration of potassium in the soluble
[10]
KIM, H.; ZEMEL, M.B. In vitro estimation of the potential
fraction to 54 and 44%, respectively, of the total amount
bioavailability of calcium from sea mustard (Undaria
of element initially present. At the same time, the fraction
pinnatifid ,
a) milk, and spinach under simulated normal
bound to insoluble ligands and the residual fraction
and reduced gastric acid conditions. Journal of Food
decreased (Figures 2 and 3 .) Losses of potassium to the
Science, v. 51, n. 4, p. 957-959, 1986.
blanching water increased from 25 to 53% with longer
[11]
NATIONAL RESEARCH COUNCIL. 1989. Recommended
blanching times.
dietary allowances. National Academy Press,
The concentration of soluble sodium in the raw
Washington, D.C. 284 p.
Ciênc. Tecnol. Aliment., Campinas, 25(3): 419-424, jul.-set. 2005
423

---

Effect of blanching on spinach mineral elements, Kawashima & Valente Soares
[12]
RAO, B.S.N. Methods for the determination of
diet and rat gastrointestinal tract contents. British
bioavailability of trace metals: a critical evaluation.
Journal of Nutrition, v. 67, p. 437- 444, 1992.
Journal of Food Science and Technolog , v
y . 31, n. 5, p.
[18]
TESSIER, A.; CAMPBELL, P.G.C.; BISSON, M.
353-361, 1994.
Sequential extraction procedure for the speciation of
[13]
RAZAGUI, I.B.; BARLOW, P.J. A chemical clean-up
particulate trace metals. Analytical Chemistry, v. 51, n.
procedure to reduce trace metal contamination from
7, p. 844-850, 1979.
laboratory blenders. Food Chemistry, v. 44, n. 4, p. 309-
[19]
WEAVER, C.M.; PLAWECKI, K.L. Dietary calcium:
312, 1992.
adequacy of a vegetarian diet. American Journal of
[14]
REYKDAL, C.; LEE, K. Soluble, dialyzabel and ionic
Clinical Nutrition, v. 59 (suppl), p. 1238-1241, 1994.
calcium in raw and processed skim milk, whole milk and
[20]
spinach. Journal of Food Science, v. 56, n. 3, p. 864-
WOLTERS, M.G.E.; DIEPENMAAT, H.B.; HERMUS,
R.J.J.; VORAGEN, A.G.J. Relation between in vitro
866, 1991.
availability of minerals and food composition: a
[15]
SAVAGE, G.P.; VANHANEN, L.; MASON, S.M.; ROSS,
mathematical model. Journal of Food Science, v. 58, n.
A.B. Effect of cooking on the soluble and insoluble
6, p. 1349-1355, 1993.
oxalate content of some New Zealand foods. Journal of
Food Composition and Analysis, v. 13, p. 201-206,
6 - ACKNOWLEDGEMENT
2000.
The authors gratefully acknowledge the
[16]
SCHMIDT, D.R.; MACDONALD, H.A.; KELLY, W.C.
financial support from the Fundação de Amparo à
Solubility of iron, calcium and magnesium in amaranth
Pesquisa do Estado de São Paulo (FAPESP) and a
and collard leaves. Qualitas Plantarum, Plant Foods for
graduate scholarship from the Coordenadoria de
Human Nutrition, v. 4, p. 373-378, 1974.
Aperfeiçoamento de Pessoal de Nível Superior
[17]
SIMPSON, R.J.; SIDHAR, S.; PETERS, T.J. Application of
(CAPES) to the first author.
selective extraction to the study of iron species present in
424
Ciênc. Tecnol. Aliment., Campinas, 25(3): 419-424, jul.-set. 2005

---