Analytical and Bioanalytical Chemistry

, Volume 381, Issue 7, pp 1460–1466

Formic acid solubilization of marine biological tissues for multi-element determination by ETAAS and ICP–AES

  • Christine Scriver
  • Masahiko Kan
  • Scott Willie
  • Catherine Soo
  • H. Birnboim
Original Paper

DOI: 10.1007/s00216-005-3125-4

Cite this article as:
Scriver, C., Kan, M., Willie, S. et al. Anal Bioanal Chem (2005) 381: 1460. doi:10.1007/s00216-005-3125-4

Abstract

A simple, fast method is described for the determination of Ag, As, Cd, Cu, Cr, Fe, Ni, and Se in marine biological tissues by electrothermal atomic-absorption spectrometry (ETAAS) and Na, Ca, K, Mg, Fe, Cu, and Mn by inductively coupled plasma–atomic emission spectrometry (ICP–AES). Solubilization of the biological tissue was achieved by using formic acid with vortex mixing followed by heating to 50°C in an ultrasonic bath. Once solubilized, the tissues were diluted to an appropriate volume with water for analysis. Aliquots were sampled into a graphite furnace and ICP–AES using a conventional autosampler. The method was validated by use of biological certified reference materials from NRC, DORM-2, DOLT-2, DOLT-3, LUTS-1, TORT-2, and NIST SRMs 1566b and 2976. Simplicity and reduced sample-preparation time prove to be the major advantages to the technique.

Keywords

Trace metals Biological tissues Formic acid ETAAS ICP–AES 

Introduction

Analytical methods used to prepare biological tissue samples for the determination of trace metals predominately involve digesting the sample with oxidizing acids using either a hot-plate or microwave heating [1, 2]. However, alternative methods such as dry ashing [3, 4], solvent extraction [5], ultrasound-assisted extraction with acids [6, 7] or enzymatic hydrolysis [8] and pressure-assisted chelating extractions [9] have been reported. Slurry and solid-sampling procedures for electrothermal vaporization atomic absorption spectrometry (ETAAS) and inductively coupled plasma atomic emission spectrometry (ICP–AES) have also gained widespread acceptance as sample-preparation techniques [10]. The slurry technique is a particularly attractive method wherein a stabilized suspension of a finely powdered sample is prepared. Slurry sampling combines the advantages of both liquid and solid sampling, the samples are easy to prepare, do not require aggressive chemical pretreatment, are less susceptible to contamination, and their use can reduce the possibility of analyte loss before analysis. Simple aqueous standards can often be used as calibrants. One aspect of the slurry technique that can create difficulty is the stabilization of the slurries. Various sample pretreatment procedures and additives have been evaluated for formation and homogenization of stable suspensions [11]. Methods utilizing ultrasonic agitation [12], gas bubbling [13], and the addition of dispersing agents such Triton X-100 [12] has been reported.

This laboratory recently investigated an alternative sample-preparation method using tetramethylammonium hydroxide (TMAH) for determination of Cu, Cd, Ni, Pb, Cr, and Mn by ETAAS [14] and total Hg by CVAAS [15] in a suite of marine biological certified reference materials (CRMs). The use of TMAH has shown to be an attractive approach for solubilizing samples before introduction into the graphite furnace [16, 17, 18] or ICP [19]. The solutions are stable and aggressive chemical pretreatment is avoided. The procedure is also less susceptible to contamination, and the possibility of analyte loss before analysis is reduced. An unfortunate negative aspect of this procedure was the unpleasant odor emitted from the solutions.

The study reported in this paper attempted to further simplify sample-preparation procedures by using formic acid to solubilize solid biological tissues. This method was validated by determination of trace metals in NRCC certified reference materials (CRM) TORT-2, DORM-2, DOLT-2, DOLT-3 and LUTS-1 and SRM NIST-2976 and NIST-1566b.

Experimental

Instrumentation

Analytical measurements were performed using a Perkin–Elmer (Norwalk, CT, USA) model 4100Z transversely heated graphite furnace atomic-absorption (THGA) spectrometer fitted with an AS-70 autosampler. Hollow-cathode lamps for Ag, Cd, Cu, Cr, Fe, and Ni (SCP Science, Montreal, Canada) and electrodeless discharge lamps (Perkin–Elmer, USA) for As and Se served as the radiation sources.

A Perkin–Elmer Optima 3000 radial view ICP–AES was used for the major elements and selected minor elements. A conventional cross-flow nebulizer with chemical-resistant gem tips and an alumina injector was used. A selection of appropriate analyte lines was used for ICP–AES determinations. Typical background corrections were applied to each analyte line.

Reagents

Inorganic standards were prepared by dissolving high purity metal, oxides or salts in appropriate acids. Standards for analysis were prepared by serial dilution of the stock solutions with 1% m/v HNO3. Pd (10,000 ppm prepared from high-purity metal), Mg(NO3)2 (Anachemia Science, Montreal, QC, Canada), and NH4H2PO4 (Alfa Aesar, Ward Hill, MA, USA) were used as modifiers. Formic acid (EM Science, Merck, Germany) was used as received. De-ionized water (DW) (Nanopure, Barnstead/Thermolyne, Boston, MA, USA) was used for all dilutions. Hydrogen peroxide (30%) (Anachemia Science) and hydrochloric and nitric acids were purified in house by sub-boiling distillation of reagent grade feedstocks using a quartz still.

CRM prepared from dogfish liver (DOLT-2, and DOLT-3), dogfish flesh (DORM-2), and lobster hepatopancreas (LUTS-1 and TORT-2) available from the National Research Council of Canada (Ottawa, Ontario, Canada) and mussel tissue (SRM 2976) and oyster tissue (SRM 1566b) from NIST (Gaithersburg, MD, USA) were used to assess the accuracy of the technique.

Sample-preparation procedure

A nominal 0.25 or 1.0 g of tissue for ETAAS and ICP–AES, respectively, was placed in a plastic centrifuge tube (or similar container) and 10 or 40 mL concentrated formic acid was added. Vortex mixing (for approximately 1–2 min) was used to aid wetting of the sample. The acidified sample mixture was then placed in an ultrasonic bath at a temperature of 50°C for 2–4 h to accelerate solubilization. Elimination of the ultrasonic agitation step resulted in a gelatine-like consistency for certain samples making reproducible pipetting difficult. Sample solutions were stored at room temperature and diluted with DW before analysis. Sample solubilization and storage were performed in the same vials simplifying the procedure by eliminating sample transfer.

For conventional acid digestion a nominal 0.25 g of tissue was weighed into Teflon digestion vessels (CEM ACV Type) and 7 mL of HNO3 and 200 μL (30%) H2O2 were added to each sample and three blanks. The vessels were capped and the samples digested using the microwave digestion system (CEM Microwave Digestion System Model MDS-2100). A typical microwave digestion program consists of a five-step power ramp to increase pressure from 20 to 120 psig over the course of 1.5 h. The samples and blanks were cooled and transferred to 25 mL final volume with DW and stored in precleaned polyethylene screw-capped bottles.

ETAAS Procedure

Furnace programs were selected to achieve gradual drying of the sample before pyrolysis and atomisation. Table 1 summarizes the furnace program for each analyte, aliquot mixtures of 10–25 μL were deposited in the furnace by using a conventional autosampler. Calibration was accomplished using an external curve with standards prepared in dilute nitric acid. Peak area measurement was used. Selection and concentration of matrix modifiers for analyte determination by ETAAS were based on Perkin–Elmer recommended conditions [20] for all analytes, with the exception of As and Se (Fig. 1).
Table 1

Operating conditions for furnace programs for ETAAS and ICP–AES

Analyte

Temperature (°C)

Pyrolysisa

Atomizationb

ETAAS operating conditions

Ag

800

1600

As

1000

2250

Cd

500

1500

Cr

800

2300

Cu

800

2100

Fe

700

2100

Ni

1100

2450

Se

1300

2100

ICP–AES operating conditions

Argon flow rates (L min−1)

Plasma

15

Auxiliary

0.5

Nebulizer

0.8

RF Power

1300 W

Sample flow rate

1.0 mL min−1

Viewing height

15 mm

Dry step−110°C, 10 s ramp, 20 s hold

Furnace clean step—2 s at 2400°C

aPyrolysis step—15 s ramp, 15 s hold

bAtomization step—0 s ramp, 5 s hold, 0 mL min−1 gas flow

Fig. 1

Char curves with (filled squares) and without (crosses) modifiers. Modifiers were: 5 μg Pd and 3 μg Mg(NO3)2 for Ag, As, Cu; 15 μg Mg(NO3)2 for Cr, Fe; 3 μg Mg(NO3)2 and 50 μg NH4H4PO4 for Cd;10 μg Pd for Se; and no modifier for Ni

ICP–AES Procedure

The instrument was optimized daily according to manufacturer recommendations. Table 1 describes the operating conditions used for analysis by ICP–AES. Mixed calibration standard solutions were prepared by combining appropriate volumes of stock solutions in volumetric flasks. A calibration blank was used to establish the calibration curve. A method blank was carried through the complete sample-preparation procedure and used to identify possible contamination resulting from reagents and to determine the limit of detection.

Results and discussion

Sample preparation

It has been recognized in the medical field that formic acid is an acceptable means of preparing select biological tissues for analysis [21]. Initial attempts using 5–10 mL dilute formic acid to solubilize the tissue samples were unsuccessful, and use of concentrated formic acid was required. Also, overnight reaction of acid and sample worked well with certain samples but other types resulted in a solution too viscous to sample. Biological samples of muscle tissues (e.g. DORM-2) were found to be the most difficult to solubilize, whereas the liver samples (DOLT-2 and DOLT-3) were more easily solubilized. Further investigation found the most consistent and rapid approach was to use vortex mixing to achieve wetting of the sample before heating in a water bath at 50°C for approximately 2–4 h with ultrasonic agitation. Use of ultrasonic agitation in the water bath increased the rate of digestion [6]. The resulting sample once treated was not viscous and was stable for several months. The CRM LUTS-1, which is packaged as a slurry containing 85% water, was also solubilized easily using this procedure.

ETAAS determination

It was necessary to daily rinse the autosampler capillary after a day of analysis to prevent build up of reduced Pd inside the capillary. Use of dilute Triton X-100 solution in the capillary-wash assisted in keeping the autosampler tubing clear. Positive displacement pipettes were found to be preferable (but not required) for accurate and precise pipeting of the solubilized tissue. Figure 1 summarizes the pyrolysis curves with and without the use of modifiers for a solubilized sample of TORT-2. A modifier mixture of 5 μg Pd and 3 μg Mg(NO3)2for Se [20] resulted in poor reproducibility; consequently, an alternative modifier (10 μg Pd) [22, 23] was successfully used for this element. Arsenic results obtained using modifiers (5 μg Pd and 3 μg Mg(NO3)2[20] were satisfactory for DORM-2, TORT-2, LUTS-1, and SRM 2976. Exceptions were SRM 1566b, and DOLT-3 and further investigation was required to achieve acceptable results. The successful modifier must be capable of normalizing sensitivity differences between the As species present in the sample and the As(V) standard. It is known the predominant form of As in these marine samples is arsenobetaine [24]. Deaker and Maher [25] have studied modifiers for determination of As in marine biological tissues and concluded 16 μg Pd and 10 μg Mg was an optimum compromise between stabilization of the As standards and the digested marine tissues. Because results were slightly more erratic for As than for other elements, it is suggested that the method of standard additions may be a conservative and preferable approach for determination of this difficult element when a variety of samples are measured.

An experiment was performed to ensure that a 1-mL sample in an autosampler cup would not partition and reproducible results could be obtained during a typical analysis of samples and standards that may last several hours. The Cu results obtained when an aliquot from the upper portion of a DORM-2 solution (diluted fivefold with DW) was sampled into the graphite furnace at 15 min intervals indicated no detectable change in the measured absorbance over several hours. An exception was found for Cr and Ni in CRM DORM-2 where a decrease in absorbance was observed over short period of time and the results were quite erratic. It is strongly suspected contamination of Cr and Ni occurred from the stainless steel blades used to break up the raw material in the preparation of these reference material samples [26]. However, the contamination was dispersed sufficiently throughout DORM-2 to enable certification. A similar situation exists for Cr in DOLT-3, where certification was not performed and an information value of 3.5 mg kg−1 is provided. Chromium values varying between 1 and 1.5 mg kg−1 in the formic acid prepared solution were found for this sample. Contamination from the grinding process is not suspected in CRMs DOLT-2, TORT-2 and LUTS-1 and the formic acid results are in agreement with the certified values. To further investigate if the Cr and Ni in DORM-2 and DOLT-3 are insoluble with formic acid treatment, a mass balance experiment was performed for both CRMs. Solubilized solutions of DORM-2 and DOLT-3 were filtered through a 0.45-μm filter so that particles of Ni and Cr were retained on the filter. The filter was then digested with aqua regia. The solubilized solution that passed through the filter and the solution from the digested filter were analyzed. Summation of the results from these analyses resulted in Cr and Ni values near the information values provided on the certificates. It is possible the formic acid-solubilized samples may represent the “true” or non-contaminated concentrations of Cr and Ni in these tissues. Further studies would be required to confirm the analyte is derived from the “natural” component in the biological tissue versus analyte that may be contributed during the processing of the powdered material.

ETAAS analytical results

Analytical results were obtained by preparing a calibration curve by use of solutions in dilute nitric acid. Samples were diluted with DW, where necessary, to fall within the linear calibration range. A standard-addition calibration was also prepared for each element to ensure the slopes of the calibration plots obtained from dilute nitric acid and formic acid solutions were similar.

Initially the analytical blank was difficult to control, because of bottle-to-bottle variation of Ni and Cr in the formic acid. Several brands were analyzed until a supplier was found who could provide bottles with low and consistent metal concentrations. Table 2 summarizes the trace metal concentration from three brands of formic acid, determined by ICP–MS. Figures of merit for analytes measured by ETAAS are outlined in Table 3. Absolute detection limits for Ag, As, Cd, Cr, Cu, Fe, Ni, and Se were estimated by replicate measurement (n=7) of the undiluted preparation blanks taken through the procedure and quantified by external calibration. Analytical results for ETAAS are summarized in Table 4. As previously discussed, the results for Ni, Cr, and Fe in DORM-2 and Ni and Cr in DOLT-3 are low, because of suspected metallic contamination during the processing of these materials. Similarly low results were also obtained for Ni and Cr in DORM-2 using TMAH solubilization [14].
Table 2

Trace metals in formic acid

Analyte

Concentrationa (ng mL−1)

Ag

0.026±0.007

As

<0.01

Cd

0.015±0.012

Cr

4.3±0.8

Cu

0.26±0.10

Fe

16.8±2.5

Ni

2.5±0.6

Se

<0.01

aAnalysis by ICP–MS

Table 3

Figures of merit for ETAAS and ICP–AES determinations

Analyte

LODa(ng mL−1)

LODb(μg g−1)

ETAAS

Ag

0.15

0.01

As

0.4

0.02

Cd

0.26

0.01

Cr

0.34

0.01

Cu

0.56

0.02

Fe

6.2

0.25

Ni

3.3

0.13

Se

3.4

0.14

ICP–AES

Na

50

100

Ca

3

6

Mg

1

2

K

100

200

Fe

3

6

Cu

2

4

Mn

0.3

1

LOD—3σ of blank (n=7)

aIn formic acid solution

bAs typical solubilization of 0.25 g tissue in 10 mL acid

Table 4

Analytical results from ETAAS

 

Ag (mg kg−1)

As (mg kg−1)

Cd (mg kg−1)

Sample

Determined*

Certified

Determined

Certified

Determined

Certified

DOLT-3

1.31±0.08

1.20±0.07

9.9±0.6

10.2±0.5

19.5±0.4

19.4±0.60

DORM-2

0.06±0.01

0.041±0.013

18.9±2.1

18.0±1.1

0.039±0.005

0.043±0.008

TORT-2

7.4±0.64

n.c.

21.2±1.3

21.6±1.8

24.5±1.81

26.7±0.60

LUTS-1

0.58±0.08

0.580±0.049

2.8±0.1

2.83±0.13

1.93±0.39

2.12±0.15

SRM 2976

<0.01

0.011±0.005

13.5±0.3

13.3±1.8

0.71±0.015

0.82±0.16

SRM 1566b

0.672±0.024

0.666±0.009

7.5±0.3

7.65±0.65

2.76±0.26

2.48±0.08

 

Cu (mg kg−1)

Fe (mg kg−1)

Se (mg kg−1)

Sample

Determined

Certified

Determined

Certified

Determined

Certified

DOLT-3

35.0±3.9

31.2±1.0

1511±10

1484±57

7.33±0.25

7.06±0.48

DORM-2

2.28±0.09

2.34±0.16

45±2

142±10

1.38±0.04

1.40±0.09

TORT-2

110±10

106±10

107±3.3

105±13

5.62±0.23

5.63±0.67

LUTS-1

16.0±1.3

15.9±1.2

11.2±0.5

11.6±0.9

0.58±0.02

0.641±0.054

SRM 2976 

4.05±0.13

4.02±0.33

181±2

171±4.9

1.89±0.06

1.80±0.15

SRM 1566b

73.7±6.0

71.6±1.6

209±9

205.8±6.8

2.04±0.08

2.06±0.15

 

Cr (mg kg−1)

Ni (mg kg−1)

Sample

Determined

Certified

Determined

Certified

DOLT-2

0.43±0.019

0.37±0.08

0.25±0.06

0.20±0.02

DOLT-3

1–2

3.5 n.c.

1.9±0.1

2.72±0.19

DORM-2

1–16

34.7±5.5

4.6±0.5

19.4±3.1

TORT-2

0.77±0.04

0.77±0.15

2.45±0.17

2.50±0.19

LUTS-1

0.072±0.001

0.079±0.012

0.17±0.01

0.20±0.034

SRM 1566b

n.d.

n.c.

0.99±0.01

1.04±0.09

     

Determined values—mean and standard deviation, n=3

Certified value—± represents uncertainty reported in certificate

n.d.—Not determined

n.c.—Not certified

Determination by ICP–AES

One aim of this study was to analyze the solubilized samples using an external calibration curve rather than relying on the method of standard additions. Because the organic matter in the samples is not destroyed in this procedure, there is the possibility of physical interferences associated with sample nebulization and transport processes; changes in viscosity and surface tension can also cause significant inaccuracies. In an effort to avoid these potential problems, the samples were diluted 50-fold with water before analysis by ICP–AES. Unfortunately, there was a matrix effect with the samples that prevented use for quantification of a simple external calibration curve obtained from analysis of nitric acid or formic acid solutions. Table 5 presents the results for the major elements Na, Ca, K, and Mg and minor elements Fe, Cu, and Mn in CRM TORT-2 using standard additions. The potential of this technique is demonstrated by accurate determination of elements with concentrations above the working limits of ETAAS; however, the analyst should use standard-additions calibration. Detection limits for the ICP–AES procedure are listed in Table 3. Further investigation using ICP–AES was not attempted, because accurate results required standard additions calibration, which was contrary to the aim of developing a simple and robust procedure.
Table 5

Analytical results for major and minor elements in CRM TORT-2 by ICP–AES

Element and wavelength (nm)

Concentration (mg g−1)

Determineda

Acid digestionb

Na 588.995

1.41±0.03

1.3±0.1

Na 589.592

1.40±0.03

Na 330.237

1.46±0.05

Ca 430.253

0.37±0.06

0.39±0.02

Ca 317.933

0.37±0.01

Ca 396.847

0.38±0.03

Mg 279.079

0.12±0.01

0.11±0.02

Mg 280.270

0.12±0.01

Mg 285.213

0.12±0.01

 

Concentration (μg g−1)

Determined

Certified

Fe 238.204

101±11a

105±13

Fe 259.940

101±10a

Cu 324.754

116±5.4c

106±10

Cu 224.700

115±7.0c

Mn 257.610

14.5±0.7a

13.6±1.2

Mn 260.569

14.7±0.6a

aMean and standard deviation n=7

bMean and standard deviation n=3

cMean and standard deviation n=6

Comparison with solubilization of tissues by use of TMAH

The use of TMAH in similar applications has provided favorable results [14, 16, 23]. As shown in this study, similar results could also be achieved using formic acid solubilization. Comparison of the two approaches reveals each procedure has minor advantages. Formic acid solubilization results in minimal sample odor compared with TMAH samples, and a full autosampler tray can be prepared and positioned on the instrument without ventilation of the samples. It was also observed, that the ETAAS furnace did not require frequent cleaning to remove deposits, as was required with the TMAH experiments. Alternatively, the TMAH reaction can proceed overnight at room temperature whereas the optimum solubilization procedure for all samples using formic acid required mixing and heat. Similar graphite tube longevity [14, 16, 27, 28] was been observed for both TMAH and formic acid. For both TMAH and formic acid degradation of the graphite tube was slower than for use of nitric acid solutions. For example, a THGA tube would typically last in excess of 700 firings before degradation of data was evident for elements requiring high pyrolysis and atomization temperatures, compared with 400–500 firings when analyzing a typical suite of elements in nitric acid-digested tissues. Both approaches provide fast, simple and inexpensive alternatives to conventional acid digestion and may find useful roles in determinations of trace metals in biological tissues.

Conclusions

The use of formic acid for solubilizing biological tissue is an alternative sample-preparation procedure for analysis by common instrumentation typically available in most analytical laboratories. The simple use of external calibration makes this technique very easily applied to analysis by ETAAS, but less desirable for ICP–AES, for which standard additions calibration must be used, because of matrix effects. Unlike sample-preparation using acid digestion in Teflon vessels or ultrasonic extraction the sample size can be varied and is not limited to a few hundred milligrams. The procedure is less aggressive than conventional acid digestion and has shown to be a useful analytical approach in speciation studies [24].

Acknowledgments

The authors thank P. Grinberg for the ICP–MS determinations.

Copyright information

© Springer-Verlag 2005

Authors and Affiliations

  • Christine Scriver
    • 1
  • Masahiko Kan
    • 2
  • Scott Willie
    • 1
  • Catherine Soo
    • 3
  • H. Birnboim
    • 3
  1. 1.Institute for National Measurement StandardsNational Research Council of CanadaOttawaCanada
  2. 2.Environmental Information Measurement SciencesHokkaido University of Education SapporoSapporoJapan
  3. 3.Centre for Cancer TherapeuticsOttawa Regional Cancer CentreOttawaCanada

Personalised recommendations