Food Chemistry 173 (2015) 682–684

Contents lists available at ScienceDirect

Food Chemistry
journal homepage: www.elsevier.com/locate/foodchem

Short communication

Comparative analysis of dry ashing and wet digestion methods

for the determination of trace and heavy metals in food samples
I.O. Akinyele a, O.S. Shokunbi b,⇑
a
University of Ibadan, Faculty of Public Health, Department of Human Nutrition, Ibadan, Nigeria
b
Babcock University, Benjamin Carson (Snr.) School of Medicine, Department of Biochemistry, Ogun State, Nigeria

a r t i c l e i n f o a b s t r a c t

Article history: This study compared the dry ashing and wet digestion methods of processing food samples for elemental
Received 7 June 2014 analysis. The concentrations of trace elements (manganese, iron, copper and zinc) and heavy metals
Received in revised form 12 October 2014 (chromium, cadmium, lead and nickel) were determined in varieties of samples classified as fruits, leafy
Accepted 17 October 2014
and fruity vegetables, tubers, legumes and cereals, obtained from Abeokuta, South-West, Nigeria. The
Available online 25 October 2014
metal concentrations were determined using atomic absorption spectrophotometer according to stan-
dard methods. The accuracy of the procedures was confirmed by spiking some samples and evaluating
Keywords:
their recoveries. The metal levels evaluated were relatively higher in the dry-ashed samples than the
Dry ashing
Wet digestion
wet-digested samples. However, the results showed non-significant differences in most data obtained
Trace elements after processing with the two methods. The dry ashing method is recommended for digestion of food
Heavy metals items in these categories because: it involves lesser amount of chemical reagents and related hazards;
Food groups it requires simple equipment and achieved better recovery.
Ó 2014 Elsevier Ltd. All rights reserved.

1. Introduction several analyses of food samples (Akinyele & Osibanjo, 1982;

Bahemuka & Mubofu, 1999; Demirel, Tuzen, Saracoglu, & Soylak,
Heavy metals are among the major contaminants of food supply 2008; Onianwa, Adetola, Iwegbue, Ojo, & Tella, 1999; Onianwa,
and may be considered the most important problem to the envi- Adeyemo, Idowu, & Ogabiela, 2001; Onianwa, Lawal, Ogunkeye, &
ronment (Zaidi, Asrar, Mansoor, & Farooqui, 2005). Such a problem Orejimi, 2000; Saracoglu et al., 2007; Tuzen, 2003). However, in
is becoming more serious all over the world especially in develop- most cases, these methods were applied separately on the samples
ing countries. Considering the level of technological advancement processed and fewer metals were evaluated. At other times when
in the developing world, it becomes very necessary to obtain they were applied together with more metals involved, fewer food
accurate and reliable data on the concentrations of elements in groups were considered. Thus, the aim of this study was to
the commonly consumed foods. Furthermore, the elemental com- contribute to the establishment of more reliable, routine method
position of food items differs from one group to another due to for the determination of manganese, iron, copper, zinc, chromium,
their different particulate nature and structural matrix. This makes cadmium, lead and nickel in five food groups (fruits, leafy and fruity
it quintessential to analyse each food group with the method that vegetables, tubers, legumes and cereals). Dry ashing and wet
will provide such accurate and reliable data. digestion methods were applied in processing the samples in an
Flame/graphite furnace atomic absorption spectrometry effort to recommend the more appropriate digestion method in
(F/GFAAS) is the most utilised instrument for the determination determining the eight metals in these food groups. The reliability
of trace and heavy metal contents of food samples in most analyti- of the methods for estimation of trace elements and heavy metals
cal laboratories. Many analytical methods including AAS for trace in the samples has been checked by spiking some samples with
element determination in food materials require decomposition metal standards and determining their recovery rates.
of the sample. The dry ashing, wet and microwave digestions are
used for sample decomposition prior to the determination of trace
elements by AAS (Saracoglu, Saygi, Uluozlu, Tuzen, & Soylak, 2007). 2. Materials and methods
The dry ashing and wet digestion methods have been applied in
2.1. Apparatus
⇑ Corresponding author. Tel.: +234 803 068 9633.
E-mail addresses: [email protected] (I.O. Akinyele), [email protected] The dry ashing was done using the Uniscope muffle furnace
(O.S. Shokunbi). (model SM 9080, Surgifriend Medicals, England), while the wet

http://dx.doi.org/10.1016/j.foodchem.2014.10.097
0308-8146/Ó 2014 Elsevier Ltd. All rights reserved.
 I.O. Akinyele, O.S. Shokunbi / Food Chemistry 173 (2015) 682–684 683

digestion was done using the Digestion System 40, 1000 heating Table 2
unit of JC Tecator. A Buck Scientific Atomic Absorption Spectropho- Comparison of the recovery rates of trace elements and heavy metals after dry ashing
and wet digestion of spiked samples.
tometer (210 VGP model, East Norwalk, Connecticut, USA) was
used for determination of the metals. The determination was Metal Method Recovery (%) Metal Method Recovery (%)
carried out in an air/acetylene flame. The operating parameters Manganese 1 98.2 ± 1.0 Chromium 1 78.0 ± 2.3
for the elements determined were set as recommended by the 2 97.8 ± 1.3 2 84.3 ± 3.9
manufacturer and they are highlighted in Table 1. Iron 1 99.9 ± 0.8* Cadmium 1 99.5 ± 1.2
2 90.3 ± 1.4 2 95.1 ± 3.6
2.2. Reagents Copper 1 97.7 ± 0.7* Lead 1 90.8 ± 2.6
2 90.4 ± 1.2 2 91.3 ± 2.7
Analytical reagent-grade chemicals were employed in the Zinc 1 97.9 ± 1.5 Nickel 1 93.5 ± 1.2
processing of all samples. HNO3 and HCl were obtained from 2 97.0 ± 2.8 2 97.2 ± 2.9*
Sigma–Aldrich (Buchs, Switzerland). Distilled de-ionised water Values are expressed as mean ± standard deviation (SD).
was used in all the experiments. All plastics and glasswares were 1
Dry ashing.
2
properly cleaned by soaking with 2 M nitric acid and rinsed Wet digestion.
*
Values are significantly higher (at P < 0.05) than the values obtained from the
thoroughly with de-ionised water before use. The standard solu-
evaluation of the same sample, using the second method.
tions of analytes for calibration procedure and spiking, obtained
from Buck Scientific, were prepared by diluting a stock solution
of 1000 ppm of the investigated elements. was carried out, as described by Onianwa et al. (2001), by spiking
and homogenising several already analysed food samples with var-
2.3. Sample collection ied amounts of standard solutions of the metals. The spiked sam-
ples were then processed the same way as other samples.
Food samples purchased from the local markets in Abeokuta, Average recoveries are as highlighted in Table 2. Each sample
South-West Nigeria were investigated in this study. The food was analysed in triplicate and results reported as mean ± standard
groups include fruits (banana, water melon, orange and apple), deviation (SD) on dry weight basis.
leafy and fruity vegetables (garden egg, tomato, onion, African
spinach (Corchorus olitorius), pumpkin leaves (Telfairia occidentalis)
and ‘sokoyokoto’ (Celosia argentea)), tubers (yam and sweet 2.4.2. Wet digestion
potato), legumes (soybean, bean and groundnut) and cereals (rice, Wet digestion of samples was performed using mixtures of
maize and wheat). Each food item was purchased by systematic acids: HNO3:HCl (3:1), by a modified method of Demirel et al.
random sampling from three to six different stores. A subsample (2008). Twenty millilitres of the acid mixture was used for each
of each was obtained for analysis. Only wholesome samples were 1 g sample digested. Each mixture was heated up to 150 °C for
included for the analysis. More so, only edible portions were 2½ h on the heating digestion block. Then the acid digest was
included, whereas bruised or rotten parts were removed. Samples allowed to cool and filtered into a 25 mL volumetric flask, using
were carefully washed in de-ionised water to remove sand and Whatman filter paper and made up to mark with de-ionised water.
other impurities. They were then drained and oven-dried at This way, organic matter was destroyed in the sample and a high
100 °C to a constant weight, taking care not to overheat the sam- concentration of the sample was achieved. The blank digests and
ples. The dried samples were homogenised using Kenwood blender spiked samples were similarly processed. The recovery study of
(model BL335, Watford, UK) and stored in clean dry air-tight poly- this analytical procedure was carried out, as described by
ethylene bags until analysis. Onianwa et al. (2001), by spiking and homogenising several
already analysed food samples with varied amounts of standard
solutions of the metals. The spiked samples were then processed
2.4. Digestion procedure
the same way as other samples. Average recoveries are as high-
lighted in Table 2. Each sample was analysed in triplicate and
2.4.1. Dry ashing
results reported as mean ± standard deviation (SD) on dry weight
A modified method of Crosby (1977) was adopted. One gram of
basis.
each dried sample was weighted into a porcelain crucible and dry-
ashed in a muffle furnace by stepwise increase of temperature up
to 500 °C within 1 h and then leaving to ash at this temperature 2.5. Statistical analyses
for additional 12 h. The residue was dissolved in 1 M nitric acid, fil-
tered into a 25 mL volumetric flask using Whatman filter paper and Data were analysed by independent student’s t-test with SPSS
made up to mark with the nitric acid (1 M). The blank digests was version 15.0 (Chicago, IL) and differences were considered statisti-
similarly processed. The recovery study of this analytical procedure cally significant at P < 0.05.

Table 1
Instrumental analytical conditions of investigated elements.

Metal Wavelength (nm) Slit (nm) Sens check (lg/ml) Linear range (lg/ml) Flame type colour
Manganese 279.5 0.7 1.25 2.50 A–A, lean/blue
Iron 248.3 0.7 2.50 5.00 A–A, lean/blue
Copper 324.8 0.7 2.00 5.00 A–A, lean/blue
Zinc 213.9 0.7 0.50 2.50 A–A, lean/blue
Chromium 357.9 0.7 2.00 5.00 A–A, rich/yellow
Cadmium 228.9 0.7 0.75 2.00 A–A, lean/blue
Lead 283.3 0.7 10.00 20.0 A–A, lean/blue
Nickel 341.5 0.2 7.00 8.00 A–A, lean/blue

Buck Scientific (2003).

684 I.O. Akinyele, O.S. Shokunbi / Food Chemistry 173 (2015) 682–684

Table 3
Concentrations (lg/g) of manganese, iron, copper, zinc, chromium, cadmium, lead and nickel in various food groups.

Food group Digestion method Mn Fe Cu Zn Cr Cd Pb Ni

Fruits Dry 6.22 ± 1.25 32.33 ± 11.33 3.44 ± 0.42 8.23 ± 3.19 0.41 ± 0.32 0.09 ± 0.04 <0.08 <0.01
Wet 5.97 ± 0.852 33.77 ± 9.36 3.27 ± 0.48 8.31 ± 3.35 0.77 ± 0.18* 0.08 ± 0.04 <0.08 <0.01
Leafy and fruity vegetables Dry 15.60 ± 4.60 50.96 ± 30.13 9.77 ± 2.32* 18.98 ± 2.92 0.68 ± 0.77 0.05 ± 0.06 <0.08 <0.01
Wet 16.27 ± 4.04 44.45 ± 30.00 8.22 ± 2.10 18.09 ± 4.76 1.04 ± 0.36 0.05 ± 0.04 <0.08 <0.01
Tubers Dry 6.58 ± 2.93 7.88 ± 0.73 4.82 ± 1.34 12.40 ± 0.82 0.11 ± 0.06 <0.01 <0.08 <0.01
Wet 7.13 ± 2.99 10.63 ± 1.40* 3.97 ± 1.44 13.10 ± 0.75 0.95 ± 0.72* <0.01 <0.08 <0.01
Legumes Dry 22.17 ± 6.96 45.75 ± 15.18 9.21 ± 1.52 38.86 ± 6.25 <0.04 <0.01 <0.08 1.13 ± 0.22
Wet 23.19 ± 6.67 41.11 ± 14.76 8.91 ± 1.41 37.81 ± 8.22 <0.04 <0.01 <0.08 2.16 ± 0.24*
Cereals Dry 11.27 ± 12.62 17.79 ± 8.74 2.84 ± 0.61 20.13 ± 7.26 <0.04 <0.01 <0.08 <0.01
Wet 9.56 ± 10.11 14.45 ± 7.70 3.18 ± 0.39 19.75 ± 6.23 <0.04 <0.01 <0.08 <0.01

Values are expressed as mean ± standard deviation (SD).

*
Values are significantly higher (at P < 0.05) than the values obtained from the evaluation of the same sample, using the second method.

3. Results and discussion manganese and iron obtained from samples processed by dry ash-
ing and wet digestion (Demirel et al., 2008). Thus, these data affirm
The average recovery rates (%) of the minerals analysed after our findings.
the dry ashing and wet digestion are presented in Table 2. The
recovery values from both dry-ashed and wet-digested samples 4. Conclusion
were nearly quantitative (>90%), except in the chromium analysis,
where the recoveries were about 80%. This did not vary signifi- In this study, we have been able to analyse food samples cover-
cantly with the food type spiked. In most of the samples spiked, ing wider scope (five food groups) with the two methods. The dry
dry-ashed samples had slightly elevated recovery rates. The recov- ashing procedure will be recommended because of four reasons: it
ery values of iron and copper from samples processed via dry ash- requires the use of smaller amount of chemicals hence cost effec-
ing were significantly higher (P < 0.05) than those of wet digestion. tive; it involves lesser risks associated with chemical usage; it
However, the recovery rate of nickel is significantly higher in wet- requires simple equipment (muffle furnace) that is easily handled;
digested samples than dry-ashed samples. The recovery rates and it achieved better recovery in the samples. Moreover, the
obtained for dry-ashed samples in this study were slightly higher obtained analytical values make this modified method more suit-
than the 92.6%, 97.3%, 94% and 91% reported respectively for Cu, able for the analysis of manganese, iron, copper, zinc, chromium,
Zn, Cd and Ni analysed from their spiked samples that were cadmium, lead and nickel in fruits, leafy and fruity vegetables,
dry-ashed (Onianwa et al., 2000, 2001). Considering our recovery tubers, legumes and cereals.
studies results, dry ashing method is somewhat more sensitive
than this wet digestion method.
The concentrations (lg/g) of the trace elements and heavy met- References
als‘ in the various food groups analysed are presented in Table 3. A Akinyele, I. O., & Osibanjo, O. (1982). Levels of some trace elements in hospital diets.
general comparison of the dry ashing and wet digestion methods Food Chemistry, 8, 247–251.
showed no statistically significant difference (P > 0.05) in all the Bahemuka, T. E., & Mubofu, E. B. (1999). Heavy metals in edible green vegetables
grown along the sites of the Sinza and Msimbazi rivers in Dar es Salaam,
results for manganese, zinc, cadmium and lead. This data suggest
Tanzania. Food Chemistry, 66, 63–66.
that either of the protocols of the dry ashing or wet digestion can Buck Scientific. (2003). Buck scientific 210/211VGP atomic absorption
be reliably applied in processing food samples in these food groups spectrophotometer operator’s manual. http://www.bucksci.com/catalogs/210-
211-Users-Manual.pdf Accessed 30.05.14.
analysed, while evaluating their manganese, zinc, cadmium and
Crosby, N. T. (1977). Determination of metals in foods: A review. The Analyst, 102,
lead contents. In most of the food groups, the concentrations of 223–268.
these minerals in the samples processed by dry ashing are slightly Demirel, S., Tuzen, M., Saracoglu, S., & Soylak, M. (2008). Evaluation of various
higher. This is further corroborated by the rates obtained from the digestion procedures for trace element contents of some food materials. Journal
of Hazardous Materials, 152, 1020–1026.
recovery studies. Onianwa, P. C., Adetola, I. G., Iwegbue, C. M. A., Ojo, M. F., & Tella, O. O. (1999). Trace
The levels of iron in the wet-digested tubers were significantly heavy metals composition of some Nigerian beverages and food drinks. Food
higher than those of the dry-ashed samples. As for copper, the con- Chemistry, 66, 275–279.
Onianwa, P. C., Adeyemo, A. O., Idowu, O. E., & Ogabiela, E. E. (2001). Copper and
centrations were significantly higher (P < 0.05) in leafy and fruity zinc contents of Nigerian foods and estimates of the adult dietary intakes. Food
vegetables dry-ashed, whereas all other samples were not signifi- Chemistry, 72, 89–95.
cantly different in concentrations. Fruits and tubers wet-digested Onianwa, P. C., Lawal, J. A., Ogunkeye, A. A., & Orejimi, B. M. (2000). Cadmium and
nickel composition of Nigerian foods. Journal of Food Composition and Analysis,
had significantly higher chromium contents. The nickel contents 13, 961–969.
were also only significantly higher in the wet-digested legumes. Saracoglu, S., Saygi, K. O., Uluozlu, O. D., Tuzen, M., & Soylak, M. (2007).
Previous data published by Saracoglu et al. (2007) showed that Determination of trace element contents of baby foods from Turkey. Food
Chemistry, 105, 280–285.
there were no significant differences in the levels of manganese,
Tuzen, M. (2003). Determination of heavy metals in soil, mushroom and plant
zinc, chromium and nickel of baby food processed via dry ashing samples by atomic absorption spectrometry. Microchemical Journal, 74,
and wet digestion methods. Reports of recovery studies using some 289–297.
Zaidi, M. I., Asrar, A., Mansoor, A., & Farooqui, M. A. (2005). The heavy metal
certified reference materials (NIST SRM 1537a tomato leaves and
concentrations along roadside trees of Quetta and its effects on public health.
NIST SRM 1515 apple leaves) and tomato sauce samples also Journal of Applied Sciences, 5(4), 708–711.
showed a non significant difference in the values of copper, zinc,