- Email: [email protected]
Evaluation of some wet digestions methods for reliable determination of total phosphorus in Australian soils
Microchemical Journal 111 (2013) 47–52
Contents lists available at SciVerse ScienceDirect
Microchemical Journal journal homepage: www.elsevier.com/locate/microc
Evaluation of some wet digestions methods for reliable determination of total phosphorus in Australian soils Benjamin Webb, Samuel B. Adeloju ⁎ NanoScience and Sensor Technology Research Group, School of Applied Sciences and Engineering, Monash University, Gippsland Campus, Churchill, Victoria 3842, Australia
a r t i c l e
i n f o
Article history: Received 30 June 2012 Received in revised form 27 January 2013 Accepted 1 February 2013 Available online 9 February 2013 Keywords: Phosphorus Total P concentration Soil Wet digestion Acid digestion
a b s t r a c t Four common wet digestion methods have been evaluated for reliable determination of total phosphorus (P) concentrations in soil samples. Wet digestion of soil samples with nitric acid gave the highest recovery of total P concentrations with a percentage recovery of 90.1±0.9% (n=3). A lower percentage recovery of 87.0±1.4% (n=3) was achieved by wet digestion of soil samples with sulphuric acid. The use of acid mixture or acid–alkaline mixture for wet digestion of soil samples gave phosphorus recoveries of 82.4±1.9% (n=3) and 85.4±2.1% (n=3) with nitric acid–sulphuric acid mixture and nitric acid–potassium persulphate mixture, respectively. Substantial improvement in phosphorus recoveries with wet digestion of soil samples with sulphuric acid was achieved by further treatment of digested soil samples with sulphuric acid, resulting in a recovery of 92.8±1.0% (n=3), which was higher than possible with other acid and mixtures. The wet digestion of soil samples with sulphuric acid was also the only method which met reactivity and safety considerations. The successful utilisation of wet digestion with sulphuric acid for reliable determination of total P concentrations in a range of soil samples from some Australian dairy and beef rearing pastoral land is reported. © 2013 Elsevier B.V. All rights reserved.
1. Introduction Phosphorus (P) is an essential nutrient in natural ecosystems and it is also critically important to agricultural systems [1–6]. It is required for many agricultural processes including photosynthesis, nitrogen fixation, flowering, seeding and fruit maturation [3–6]. Soils in most parts of the world are low in P and deficient for optimal agricultural production [1,4–8]. This deficiency is often compensated for by applying various sources of P to soil, and thereby increasing the P status of the soil and its agricultural yield. Unfortunately, this mode of addition of P to soils has also had severe and detrimental effects on several waterways around the world, leading to an increased rate of eutrophication [9–12] and common presence of algal blooms which have been associated with various problems in fish, livestock and human health [4,12–14]. The presence of phosphorus (as phosphate) and nitrogen (as nitrate) has also been implicated in many cases with several reported outbreaks of blue-green algae which can have devastating effects on waterways. A very good indication of the soil P pools and management practices that may contribute to P enrichment of runoff and waterways can be obtained by conducting soil P measurements [5,6,12,15–17]. These measurements provide a basis for matching P inputs and agricultural crop demands [1,5,6,18]. Soil P measurements have been used to assess both environmental and agronomic impacts of P concentration extractable from soil [6,18–21]. ⁎ Corresponding author. Tel.: +61 3 9902 6450; fax: +61 3 9902 6738. E-mail address: [email protected] (S.B. Adeloju). 0026-265X/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.microc.2013.02.001
Total P concentration in soil is an important parameter which accounts for all forms of P within the soil. This parameter is often used to determine soil P status for phosphorus based fertiliser application and for estimation of P exports or agricultural yield [5,6,22]. It also provides a useful indication of the overall and potential nutrient supply of P, and has been used in relationship comparisons with other soil measurements. However, total P measurement is limited in that it does not differentiate between plant available P and non-available sources, such as organically bound from insoluble mineral P [23]. Nevertheless, it is still a very significant parameter for both environmental and agronomic considerations. The reliable determination of total P concentrations in soil is however not easy. It requires adequate and effective decomposition of the soil matrix to ensure complete release of P into solution prior to analytical measurement. A number of digestion methods have been proposed for the determination of total P concentrations in soil. In one method, Tan [24] used a fluoro-boric acid digestion which employed a specialised bomb digestion vessels and hydrofluoric acid. The original version of this method [25] also required specialised platinum crucibles and utensils for sodium hydroxide or sodium carbonate fusion [24]. Olsen and Sommers [26] also proposed a sodium carbonate fusion method, and this again required specialised platinum equipment. Another suggestion from Olsen and Sommers [26] involved wet digestion of soil with a perchloric acid. This method was not only complex, but had a range of safety issues and required a specialised perchloric acid fumehood. For these reasons, this digestion method is rarely used, except in specialised laboratories with adequate perchloric acid fumehood. Due to safety concern, some of the available soil method handbooks have
48
B. Webb, S.B. Adeloju / Microchemical Journal 111 (2013) 47–52
deliberately not included methods that employ perchloric acid [23]. However, it is important to note that safety consideration is not only limited to the use of perchloric acid, but also to the reactivity of some acids and/or mixtures to soil samples. In another study, Dick and Tabatabai [27] reported on the use of a sodium hypobromite/sodium hydroxide solution for total P determination in soil samples. The mixture was heated to dryness on a sand bath at 260 °C, followed by addition of formic acid and sulphuric acid. Bowman [28] also used ordered additions of sulphuric acid, hydrogen peroxide and hydrofluoric acid to decompose soil samples for total P determination. Both of these methods require considerable manipulation of the sample and are labour intensive compared to other methods [29]. The examples cited above highlight the current state of play with the determination of total P concentrations in soil. This is obviously not ideal for such a significant parameter which is highly important in assessing the inputs of P into farmland and waterways. In general, most of the currently available methods require specialised laboratory conditions or equipment, dangerous chemicals and intensive labour. To address this issue and ensure ease of attainment of reliable total P concentrations, there is a need for the development of a simpler soil digestion method that can be readily employed within standard laboratory conditions without safety concern, complex and laborious processes. This paper reports on a thorough evaluation and assessment of four wet digestion methods carried out to enable identification and/ or development of a simple and direct approach for pre-treatment of soil samples for reliable determination of total P concentrations. The wet digestion methods considered are those that are readily used for determination of other substances and that require the use of simple laboratory equipment, glassware and reagents [30,31]. The adequacy, efficiency and choice of the digestion methods were assessed on the basis of recovery efficiencies for total P concentration in soil samples, the ease of use, reproducibility and safety consideration. The application of the chosen method to the reliable determination of total P concentration in a wide range of soil samples from two intensive agricultural areas was also considered. 2. Materials and methods 2.1. Reagents and standard solutions All acids and reagents used were of analytical reagent grade solutions. All solutions were prepared and diluted with distilled deionised water (18 MΩ cm, Millipore, MA, USA). Nitric acid (0.01 M) used for adjusting the pH was prepared by diluting concentrated nitric acid with distilled deionised water. Stock phosphate solution (1000 mg P/L) was prepared by dissolving 4.393 g of potassium di-hydrogen orthophosphate in 1 L of distilled deionised water. A standard phosphate solution (100 mg P/L) was prepared by diluting an aliquot of this solution further with distilled deionised water. 2.2. Instrumentation All phosphate analyses were carried out by using an adapted method (SmartChem 140 Method 420-3651) modified for use with soil samples on the Westco SmartChem 140 automated wet chemistry discrete analyser (Westco Scientific Instruments, Inc., Brookfield, CT, USA). This method utilises an antimony–phospho-molybdate complex formed through the reaction of ammonium molybdate, antimony potassium tartate and dilute phosphorus solutions in an acid medium. Ascorbic acid is added to reduce the complex to produce a blue coloured complex measured at 880 nm. The resulting absorbance increased proportionally to P concentration in solution. The normal sample and reaction diluant in the method was deionised water, but was changed in this study to 0.01 M nitric acid to match the acidity of diluted sample digest.
2.3. Glassware All glassware and other containers were washed, soaked in a 2 M nitric acid for at least 7 days, rinsed three times with deionised water, soaked in deionised water and finally soaked in 0.1 M hydrochloric acid (HCl) until ready for use. 2.4. Heating sources Standard laboratory hotplates with aluminium surfaces were used for heating samples. The maximum temperature setting was used for each of the methods or as safety permitted. The temperatures of the extracts were recorded and their significance for sample digestion is discussed later. 2.5. Sample collection and preparation Soil samples were collected from an irrigation bay (width 30 m, length 300 m) at the Macalister Research Farm (38°00′S 146°54′E), a dairy farm situated in the Macalister Irrigation District of south-east Victoria (Australia). The soil was a natric grey Sodosol [32] and carried pastures that contained perennial ryegrass (Lolium perenne), white clover (Trifolium repens) and assorted invasive species including dock (Rumex spp.) and distichum (Paspalum paspaloides) [11,33]. 200 samples of 20 mm cores and 30 samples of 100 mm cores were collected from the sampling sites using a grid pattern. The soil cores were bulked for each depth to provide a composite sample for the sampling location. After collection, the soil was stored (4 °C) in polyurethane bag and transported to the laboratory. The bulked soil cores were air dried (40 °C), ground and passed through a 2 mm sieve. Samples were then stored in polyethylene containers at ca. 20 °C prior to analysis. Other soil samples were collected from selected agricultural sites within south-east Victoria, from 2 areas known as Maffra and Warragul. These areas were selected as geographically close agricultural areas (approximately 120 km range) with varying agricultural management practices, particularly irrigation application. A total of 14 sites were sampled, seven from each of the two areas. Soils were classified using an Australian soil classification system. The 7 sites in Warragul were of three different soil classifications, as indicated later in the results. The 7 sites in Maffra were also of 3 different soil classifications, as indicated later in the results. 2.6. Soil moisture content The moisture content was determined by heating the soil samples in a drying oven at 105 °C, cooled in desiccators and weighed repeatedly until a constant mass was obtained. The total P concentrations for each sample replicate were corrected for soil moisture content. 2.7. Digestion methods Four wet digestion methods were investigated for reliable determination of total P concentrations in soil samples. These methods were adapted from those reported previously by Adeloju et al. [30,31] for trace metal analyses. Digestion of each sample was carried out in triplicate. The specific soil digestion procedures for each of the wet digestion methods are described as follows: Method A: Nitric acid (HNO3) digestion: 0.3 g of soil sample was accurately weighed into a 100 mL Erlenmeyer flask. The flask containing sample was then transferred to a fumehood, where 10 mL of HNO3 (70%) was added and a glass funnel was inserted into the neck of the flask. The mixture was heated on a hotplate to approximately 125 °C where nitrogen oxide fumes were evolved and the volume of the mixture was reduced to approximately 2 mL. The flask was then
B. Webb, S.B. Adeloju / Microchemical Journal 111 (2013) 47–52
allowed to cool for at least 3 min. This process was repeated twice with another addition of 10 mL HNO3 each time. With the final nitric acid addition, heating was continued at 125 °C until no further nitrogen oxide fumes were given off. The funnel was then rinsed with a small volume of deionised water into the flask and the contents were allowed to cool to room temperature. The contents were transferred to a 25 mL volumetric flask and made up to the mark with deionised water. Samples were allowed to settle overnight, and an aliquot was taken by pipetting for analysis or storage. Samples were then stored at −20 °C in acid washed polyethylene containers until ready for processing. Method B: Sulphuric acid (H2SO4) digestion: 0.3 g of soil sample was accurately weighed into a 100 mL Erlenmeyer flask. The flask containing sample was then transferred to a fumehood, where 10 mL of H2SO4 (98%) was added and a glass funnel was inserted into the neck of the flask. The mixture was heated on a hotplate to approximately 160 °C until sulphite fumes were evolved and the volume of the mixture was reduced to approximately 2 mL. The flask was then allowed to cool for at least 3 min. This process was repeated 4 more times with 1 mL H2SO4 addition each time. With the final addition, heating was continued at 160 °C until no further sulphite mist was given off. The funnel was then rinsed with a small volume of deionised water into the flask and the contents were allowed to cool to room temperature. The contents were transferred to a 25 mL volumetric flask and, the digestion flask was rinsed carefully by using deionised water, and the content of the volumetric flask was mixed to avoid an acid–water inversion layer. The content of the flask was allowed to cool before being made up to the mark with deionised water. Samples were allowed to settle overnight, and an aliquot was taken by pipetting for analysis or storage. Samples were then stored at −20 °C in acid washed polyethylene containers until ready for processing. Method C: Mixture of nitric and sulphuric acids (HNO3–H2SO4) digestion: 0.3 g of soil sample was accurately weighed into a 100 mL Erlenmeyer flask. The flask containing sample was then transferred to a fumehood, where 10 mL of HNO3 (70%) and 1 mL H2SO4 (98%) were added and a glass funnel was inserted into the neck of the flask. The mixture was heated on a hotplate to approximately 130 °C where nitrogen oxide fumes were evolved and the volume of the mixture was reduced to approximately 2 mL. The flask was then allowed to cool for at least 3 min. This process was repeated 3 times with 10 mL HNO3 addition each time. With the final addition, heating was continued at 130 °C until no further nitrogen oxide fumes were given off. The funnel was then rinsed with a small volume of deionised water into the flask and the contents were allowed to cool to room temperature. The contents of the Erlenmeyer flask were transferred into a 25 mL volumetric flask and made up to the mark with deionised water. Samples were allowed to settle overnight, and an aliquot was taken by pipetting for analysis or storage. Samples were then stored at −20 °C in acid washed polyethylene containers until ready for processing. Method D: Mixture of nitric acid and potassium persulphate (HNO3– K2S2O8) digestion: 0.3 g of soil sample was accurately weighed into a 100 mL Erlenmeyer flask. The flask containing sample was then transferred to a fumehood, where 10 mL of concentrated HNO3 (70%) and 4 mL K2S2O8 (10% m/V) were added and a glass funnel was inserted into the neck of the flask. The mixture was heated on a hotplate to approximately 120 °C where nitrogen oxide fumes were evolved and the volume of the mixture was reduced to approximately 2 mL. The flask was then allowed to cool for at least
49
3 min. This process was repeated with 10 mL HNO3 addition each time. With the final addition, heating was continued at 120 °C until no further nitrogen oxide fumes were given off. The funnel was then rinsed with a small volume of deionised water into the flask and the contents were allowed to cool to room temperature. The contents of the Erlenmeyer flask were transferred into a 25 mL volumetric flask and made up to the mark with deionised water. Samples were allowed to settle overnight, and an aliquot was taken by pipetting for analysis or storage. Samples were stored at −20 °C in acid washed polyethylene containers until ready for processing. 2.8. Recovery spikes For each digestion method, a series of standard spikes were employed to assess recovery efficiency. The spike additions employed were: 0 (blank), 500, 1000 and 1500 mg P/kg. The spiked amount was added to individual soil samples prior to the addition of the digestion acid or mixture. Each sample spikes were also repeated in triplicate. 2.9. Sample preparation and calibration A 2 mL aliquot of each sample digest was diluted in a 25 mL volumetric flask with 0.01 M HNO3 solution prior to analysis. The resultant solution was then analysed for P concentration by using the SmartChem 140 discrete analyser. The results of the replicate samples were then combined and analysed statistically. Calibration was carried out using a six point calibration with 0.0, 0.5, 1.0, 2.0, 3.1 and 5.0 mg P/L. The SmartChem 140 discrete analyser automated system prepared a linear calibration plot through the dilution of a 100 mg P/L stock solution. The R 2 values for the calibration plots ranged from 0.9841 to 0.9922. 2.10. Method/reagent blanks Blank digestions of reagents were prepared for each of the 4 wet digestion methods. The same digestion procedure outlined for each digestion method was followed for each blank without the addition of soil sample. Final concentrations were corrected for the blank solution P concentrations for each method. The average blank contribution of the digestion solutions from each method were: method A (0.19 ± 0.04%), method B (0.20 ± 0.04%), method C (0.15 ± 0.02%) and method D (0.20 ± 0.05%). Evidently, there were no significant differences between the blank corrections for the four wet digestion methods. 2.11. Working environment All sample preparation and analyses were conducted in controlled laboratory conditions at 22.0 ± 0.5 °C. All digestions were undertaken in standard laboratory fume hoods. 3. Results and discussion 3.1. Recovery study As a first step in assessing the effectiveness of the four chosen wet digestion methods for reliable determination of total P concentrations in soil, their recovery efficiencies were compared and evaluated. As the two broad type of soil samples collected from the sampling sites in this study are “clay loam” and “sandy loam”, we chose to use a clay loam for the comparison because the soil samples collected from 13 of the 14 sampling sites were clay loam of different colour variations. It was not expected that the colour variation will alter
50
B. Webb, S.B. Adeloju / Microchemical Journal 111 (2013) 47–52
the ability to acid digest the clay loam samples. The chosen digestion method will be subsequently applied to the sandy loam sample. Soil samples digested with only concentrated HNO3 (method A) gave the highest percentage recovery for total P concentration, averaging 90.1± 0.9%. On the other hand, soil samples digested with only concentrated sulphuric acid (method B) gave a slightly lower recovery of 87.0± 1.4%. Contrary to expectation, the use of a mixture of concentrated nitric and sulphuric acids (method C) and a mixture of concentrated nitric acid and potassium persulphate (method D) gave even lower recoveries of 82.4± 1.9% and 85.4 ±2.1%, respectively. In particular, the lowest recovery efficiency achieved with method C was somewhat surprising. The expectation was that the increased strength of the two strong acid mixtures would enable better decomposition of the soil matrix and, at least, produce similar or better percentage recovery of total P concentration compared to that obtained with the use of only concentrated nitric acid. To clarify the above unusual trend of the recovery efficiencies obtained for the four wet digestion methods, further investigation was conducted with unspiked soil samples. Interestingly, when these four digestion methods were applied to the determination of total P concentrations in unspiked soil samples, the data in Table 1 revealed that the phosphate concentrations found increased in the following order: Method B >
Method C > Method A > Method D:
This trend suggests that the use or inclusion of H2SO4 may be beneficial for more effective release of organically bound P in soil. By taking the total P concentration obtained for the soil sample with method B as optimum (100%), the estimated recoveries of P in the unspiked soil samples for the three other methods were 93.1% (method A), 95.9% (method C) and 76.1% (method D). Based on this assumption, a notable but significant observation is that the application of methods B and C to spiked soil samples (with 87.0 ± 1.4% and 82.4 ±1.9% recoveries for methods B and C, respectively) results in a net decrease of 13–14% in each case. As the presence of H2SO4 is a common factor in both digestion methods, this observation seems to suggest that the lower recoveries obtained in the spiked soil samples may be due to an impact of this acid on the inorganic P spike. A possible link that can be deduced from this observation is that the much higher digestion temperature achieved with H2SO4 only (method B) or as a mixture (method C) may result in losses of the inorganic P spikes during soil digestion. As can be clearly seen in Table 1, the digestion methods which include
Table 1 Recovery efficiency of wet digestion methods for total P concentrations in a grey clay loam Dermosol. Sample description Method Method Method Method Method Method Method Method Method Method Method Method Method Method Method Method a b c d
A unspikeda A spike 1a A spike 2a A spike 3a B unspikedb B spike 1b B spike 2b B spike 3b C unspikedc C spike 1c C spike 2c C spike 3c D unspikedd D spike 1d D spike 2d D spike 3d
Max digestion temp. °C
P spike conc. (mg P/kg)
Total P (mg P/kg)
Recovery efficiency (%)
125
0 500 1000 1500 0 500 1000 1500 0 500 1000 1500 0 500 1000 1500
1777 ± 56 2229 ± 113 2683 ± 133 3111 ± 59 1908 ± 86 2346 ± 62 2788 ± 68 3189 ± 55 1829 ± 62 2249 ± 76 2656 ± 147 3034 ± 89 1452 ± 23 1893 ± 36 2296 ± 68 2720 ± 44
– 90.5 90.7 89.0 – 87.5 88.0 85.4 – 84.0 82.7 80.3 – 88.1 84.4 84.5
160
130
120
Concentrated HNO3 (3 × 10 mL) additions only (n = 3). Concentrated H2SO4 (1 × 10 mL and 4 × 1 mL) additions only (n = 3). Concentrated HNO3 (4×10 mL) and concentrated H2SO4 (1×1 mL) additions (n=3). Concentrated HNO3 (3×10 mL) and 10% w/v K2S2O8 (1×4 mL) additions (n=3).
H2SO4 (methods B and C) achieved higher solution temperatures, up to 40 °C higher than with other two digestion methods. Evidently, this is why method B gave the highest phosphate concentrations in the unspiked soil samples. Interestingly, the achieved solution temperature with the four digestion methods increased in the following order: Method B > Method C > Method A > Method D which correlated well with the trend observed for the total P concentrations in the unspiked soil samples. Notwithstanding the observed differences in the recovery efficiencies, the total P concentrations found in the unspiked soil samples within 0–20 mm depth by the four different digestion methods were comparatively close to those previously obtained at the same farm in 2001 (1528 mg P/kg) and 2004 (1432 mg P/kg) [10,11]. The higher concentrations found with methods A, B and C may be attributed to increasing P inputs into the sampled paddock between 2004 and 2007. 3.2. Choice of digestion method Apart from the recovery efficiencies achieved with the four wet digestion methods for total P concentrations in spiked and unspiked soil samples, proper selection of a digestion method must also consider relevant and/or associated safety issues. One of the important considerations in proposing a digestion method that can be readily employed in most laboratories for the determination of total P concentration in soils is the degree and extent of reactivity caused by the digestion medium or mixture during the digestion process. Highly reactivity digestion process can result in unexpected losses of samples and, in extreme cases, can result in accidental acid splash, corrosive reactions or other undesirable consequences. It is, therefore, highly essential that the chosen digestion method for soil samples is moderately reactive, while still ensuring complete recovery of total P concentrations in soils. The three wet digestions methods (methods A, C and D) which employed HNO3, either alone or as a mixture, were found to be very reactive, causing a sudden release of pressure when heated to the maximum temperatures. A pressure build-up below the fine particulate of the soil samples occur during digestion and led to the sudden pressure release. This can pose a serious safety concern as the digest can escape in some cases from the heated flask. This was particularly a problem with method A which used only HNO3 for soil digestion. In comparison, the pressure release and associated popping effect was somewhat subdued in method C and this appears to be due to the inclusion of H2SO4 in the mixture at the initial stage. However, more popping effect was observed with the required subsequent additions of HNO3 during digestion with this method. As a precaution, irrespective of the HNO3 based digestion method, care must be exercised and the heating temperature may need to be reduced to minimise potential hazard, but this will lead to a considerable extension of the digestion time. In contrast, no pressure build-up or popping effect was observed for sample digestion with method B, as the boiling temperature of H2SO4 was much higher. On the basis of recovery efficiency achieved for total P concentrations in unspiked soil samples and safety considerations, method B was chosen as the best of the four wet digestion methods for further investigation. Even with spiked soil samples, the recoveries obtained by this method was only slightly lower than those obtained with method A and were higher than those of methods C and D. Possible improvement in the recovery efficiency achieved with method B for determination of total P concentrations was further investigated by examining the effect of additional treatment with H2SO4. 3.3. Extended H2SO4 digestion (method B +) The possibility of improving the recovery efficiency achieved for spiked soil sample with method B was investigated by subsequent treatment of the initial sample digest with several 1 mL H2SO4 additions.
B. Webb, S.B. Adeloju / Microchemical Journal 111 (2013) 47–52
Table 2 Influence of increasing addition of H2SO4 on recovery efficiency of total P in a grey sandy loam Dermosol. Sample description Method Method Method Method Method Method Method Method Method Method Method Method a
B5 B5 B5 B5 B6 B6 B6 B6 B7 B7 B7 B7
unspiked spike 1 spike 2 spike 3 unspiked spike 1 spike 2 spike 3 unspiked spike 1 spike 2 spike 3
Additionsa (1 mL H2SO4)
Spike conc. (mg P/kg)
Total P (mg P/kg)
Recovery efficiency (%)
5 5 5 5 6 6 6 6 7 7 7 7
0 500 1000 1500 0 500 1000 1500 0 500 1000 1500
1083 ± 15 1539 ± 32 2000 ± 37 2449 ± 12 1092 ± 18 1549 ± 38 2018 ± 16 2468 ± 20 1123 ± 13 1585 ± 46 2062 ± 38 2505 ± 19
– 91.3 91.8 91.1 – 91.4 92.6 91.7 – 92.3 93.9 92.1
Please note: 0–100 mm depth samples (n = 3) were used.
96
Recovery Efficiency (%)
Method B, as described in materials and methods, involved digestion of soil sample in 10 mL concentrated H2SO4 followed by four 1 mL additions. Following exactly the same procedure, the effect of increasing the 1 mL H2SO4 additions beyond four on the recovery efficiency for total P concentration from spiked samples was of particular interest. The new methods (B5, B6, and B7 listed in Table 2 with 5, 6 and 7 additions of 1 mL H2SO4, respectively) were carried out on soil samples collected from 0 to 100 mm depth instead of the 0–20 mm depth used for comparison of the four wet digestion methods. For this reason, it was obvious that the total P concentrations obtained for these unspiked soil samples were lower than those reported in Table 1. The data in Table 2 show that the total P concentrations in the unspiked soil samples increased slightly with each 1 mL H2SO4 addition and treatment, indicating small improvement in the release of organically bound P. Also the percentage recoveries obtained for the spiked soil samples improved slightly with each acid addition. Evidently, the average spike recovery obtained increased to 91.4 ± 0.4% for method B5, to 91.9 ± 0.6% for method B 6, and further to 92.8 ± 1.0% for method B7. Fig. 1 shows the effect of increasing addition of H2SO4 on the average recoveries of total P concentration in the spiked soil samples. These results demonstrate that a net improvement in recovery efficiency of 4– 6% was achieved by further treatment of sample digest with 1–3 additions of 1 mL H2SO4. Furthermore, it is worth noting that the average spike recoveries obtained with methods B5, B 6, and B7 were all higher than that of method A. Fig. 1 also shows that the recovery of P with this approach increased considerably between the fourth and fifth 1 mL additions and then levelled out with the sixth and seventh additions. This observation clearly suggests that further acid additions will not lead to more improvement in recovery and was therefore not considered. These results also suggest that the lower P recoveries obtained earlier for the spiked soil samples collected from 0 to 20 mm depth may be due to other soil composition. The 0–20 mm soil is likely to contain higher amounts of organic material than those from 0 to 100 mm depth. This could mean that the soil samples collected from 0 to 20 mm depth were more difficult to digest. Interestingly, recovery study of spike 0–100 mm depth soil samples by the original method B digestion method gave an average recovery comparable to the original results obtained with the 0–20 mm depth samples (87.0% ± 1.4% for 0–20 mm depth versus 87.2% ± 0.9% for 0– 100 mm depth). However, it is recommended, based on the results obtained in this section that method B should be employed with at least five 1 mL H2SO4 additions. Table 3 shows that this approach was successfully employed for the determination of total P concentration in soil samples collected from two intensive agricultural areas. Evidently, the total P concentrations in these soil samples are influenced by factors such as the soil characteristics, quantity, frequency and length of P fertiliser application.
51
94 92 90 88 86 84 3
4
5
6
7
Number of 1 mL H2SO4 Additions Fig. 1. Average spike recovery (%) versus increasing number of 1 mL H2SO4 additions for wet digestion of grey sandy loam Dermosol with method B. n = 3.
4. Conclusions The careful evaluation of the four wet digestion methods for reliable determination of total phosphorus (P) concentrations in soil samples has demonstrated that the criteria for selecting an appropriate method must include safety consideration, in addition to the achievable recovery efficiency. Wet digestion with H2SO4 was found to be less reactive, most effective and safe for total P determination in soil. Further treatment of pre-digested soil samples with H2SO4 led to substantial improvement in phosphorus recovery, achieving a percentage recovery of 92.8 ±1.0%. This digestion method was successfully utilised for the reliable determination of total phosphorus concentrations in various soil samples from dairy and beef rearing pastoral areas.
Acknowledgements One of the authors (Ben Webb) wishes to acknowledge the provision of an Australian Postgraduate Research Scholarship through Monash Research Graduate School (MRGS) and other support provided by the School of Applied Sciences and Engineering. The authors also acknowledge the research funding provided for this project by Dairy Australia.
Table 3 Total phosphorus concentrations in soil samples collected from intensive agricultural areas in Warragul and Maffra (S-E Victoria, Australia). Sample site
Total P (mg P/kg)
Soil characteristica
Period post fert. appln. (day)
Last fert. appln. (mg P/ha)
W1 W2 W3 W4 W5 W6 W7 M1 M2 M3 M4 M5 M6 M7
1838 1941 3009 2680 1779 2062 1103 1936 2381 1447 1470 1775 2000 1743
GCLD GCLD RCLF RCLF RCLF GCLD GSLD GRCL GRCL GRCL GRCL GYCL GCLS GCLS
365 365 365 365 45b 20 730+ 365 45 1000+ 730+ 60 130 130
24.8 24.8 24.8 24.8 3.7b 14.8 n/a 11.8 14.5 n/a n/a 13.2 7.7 7.7
n/a = not available. a Soil description; GCLD=grey clay loam Dermosol, RCLF=red clay loam Ferrosol, GSLD=grey sandy loam Dermosol, GRCLS=grey/red clay loam Sodosol, GRCLS=grey/ yellow clay loam Sodosol, GCLS=grey clay loam Sodosol. b Site W5 has only received a single fertiliser application (3.7 kg P/ha) in three years prior to sampling.
52
B. Webb, S.B. Adeloju / Microchemical Journal 111 (2013) 47–52
References [1] P.W. Moody, M.D.A. Bolland, in: K.I. Peverill, L.A. Sparrow, D.J. Reuter (Eds.), Soils Analysis: An Interpretation Manual, CSIRO Publishing, Collingwood, Australia, 1999, p. 187. [2] R.W. McDowell, B.J.F. Biggs, A.N. Sharpley, L. Nguyen, Connecting phosphorus loss from agricultural landscapes to surface water quality, Chem. Ecol. 20 (2004) 1–40. [3] N.C. Brady, R.R. Weil, The Nature and Properties of Soils, 14th Revised ed. Pearson Prentice Hall, New Jersey, USA, 2008. [4] D. Cordell, J.O. Drangert, S. White, The story of phosphorus: global food security and food for thought, Global Environ. Change 19 (2009) 292–305. [5] J. Shen, L. Yuan, J. Zhang, H. Li, Z. Bai, X. Chen, W. Zhang, F. Zhang, Phosphorus dynamics: from soil to plant, Plant Physiol. 156 (2011) 997–1005. [6] K.W. King, J.C. Balogh, S.G. Agrawal, C.J. Tritabaugh, J.A. Ryan, Phosphorus concentration and loading reductions following changes in fertilizer application and formulation on managed turf, J. Environ. Monit. 14 (2012) 2929–2938. [7] In: D.J. Connor, D.F. Smith (Eds.), Agriculture in Victoria, Australian Institute of Agricultural Science, Parkville, Australia, 1987. [8] I.C.R. Holford, Soil phosphorus: its measurement and its uptake by plants, Aust. J. Soil Res. 35 (1997) 227–239. [9] R.B. Grayson, K.S. Tan, A. Western, Estimation of Sediment and Nutrient Loads into the Gippsland Lakes, CSIRO and University of Melbourne, Melbourne, Australia, 2001, p. 77. [10] D. Nash, B. Webb, M. Hannah, S. Adeloju, M. Toifl, K. Barlow, F. Robertson, F. Roddick, N. Porter, Changes in nitrogen and phosphorus concentrations in soil, soil water and surface run-off following grading of irrigation bays used for intensive grazing, Soil Use Manage. 23 (2007) 374–383. [11] D. Nash, M. Hannah, K. Barlow, F. Robertson, N. Mathers, C. Butler, J. Horton, A comparison of some surface soil phosphorus tests that could be used to assess P export potential, Aust. J. Soil Res. 45 (2007) 397–400. [12] In: E.K. Bünemann, A. Oberson, E. Frossard (Eds.), Phosphorus in Action: Biological Processes in Soil Phosphorus Cycling, 1st edition, Springer, 2011(XV). [13] A.M. Ebeling, L.R. Cooperband, L.G. Bundy, Phosphorus source effects on soil test phosphorus and forms of phosphorus in soil, Commun. Soil Sci. Plant Anal. 34 (2003) 1897–1917. [14] R. McDowell, A. Sharpley, P. Kleinman, Surface water quality and phosphorus applications, Encycl. Water Sci. 1 (2003) 961–964. [15] T.M. Aye, M.L. Nguyen, N.S. Bolan, M.J. Hedley, Phosphorus in soils of riparian and non-riparian wetland and buffer strips in the Waikato area, New Zealand, N.Z. J. Agric. Res. 49 (2006) 349–358. [16] A.N. Sharpley, S.J. Smith, Prediction of soluble phosphorus transport in agricultural runoff, J. Environ. Qual. 18 (1989) 313–316.
[17] A.N. Sharpley, R.W. Tillman, J.K. Syers, Use of laboratory extraction data to predict losses of dissolved inorganic phosphate in surface water and tile drainage, J. Environ. Qual. 6 (1977) 33–36. [18] N. Devau, E. Le Cadre, P. Hinsinger, F. Gérard, A mechanistic model for understanding root-induced chemical changes controlling phosphorus availability, Ann. Bot. (Lond) 105 (2010) 1183–1197. [19] A.M. Atia, A.P. Mallarino, Agronomic and environmental soil phosphorus testing in soils receiving liquid swine manure, Soil Sci. Soc. Am. J. 66 (2002) 1696–1705. [20] A.N. Sharpley, Soil phosphorus dynamics: agronomic and environmental impacts, Ecol. Eng. 5 (1995) 261–279. [21] G.M. Pierzynski, J.T. Sims, G.F. Vance, Soils and Environmental Quality, CRC Press, Boca Raton, Fl, USA, 2005. [22] D. Kara, C. Özsavasci, M. Alkan, Investigation of suitable digestion methods for the determination of total phosphorus in soils, Talanta 44 (1997) 2027–2032. [23] G.E. Rayment, F.R. Higginson, Australian Laboratory Handbook of Soil and Water Chemical Methods, Inkata Press, Melbourne, Australia, 1992. [24] K.H. Tan, Soil Sampling, Preparation and Analysis, 2nd ed. CRC Press, Boca Raton, Fl, USA, 2005. [25] B. Bernas, A new method for decomposition and comprehensive analysis of silicates by atomic absorption spectrometry, Anal. Chem. 39 (1968) 1632–1636. [26] S.R. Olsen, L.E. Sommers, in: A.L. Page, R.H. Miller, D.R. Keeney (Eds.), Methods of Soil Analysis, Part 2. Agronomy No. 9, American Society of Agronomy, Madison, WI, USA, 1982, pp. 403–430. [27] W.A. Dick, M.A. Tabatabai, Determination of orthophosphate in aqueous solutions containing labile organic and inorganic phosphorus compounds, J. Environ. Qual. 6 (1977) 82–85. [28] R. Bowman, A rapid method to determine total phosphorus in soils, Soil Sci. Soc. Am. J. 52 (1988) 1301. [29] I.P. O'Halloran, in: M.R. Carter (Ed.), Soil Sampling and Methods of Analysis, Lewis Publishers, Boca Raton, USA, 1993, pp. 213–230. [30] S.B. Adeloju, Comparison of some wet digestion and dry ashing methods for voltammetric trace element analysis, Analyst 114 (1989) 455–461. [31] S.B. Adeloju, A.M. Bond, M.H. Briggs, Critical evaluation of some wet digestion methods for the stripping voltammetric determination of selenium in biological materials, Anal. Chem. 56 (1984) 2397–2401. [32] R.F. Isbell, Australian Soil Classification, CSIRO Publishing, Melbourne, Australia, 1996. [33] B. Webb, D. Nash, M. Hannah, S. Adeloju, M. Toifl, F. Roddick, N. Porter, in: B. Singh (Ed.), Phosphorus Between Soil, Soil Water and Overland Flow for Established and Laser Graded, Border-check Irrigation Systems, SuperSoil 2004: Program and Abstracts for the 3rd Australian New Zealand Soils Conference, The Regional Institute Ltd, University of Sydney, Australia, 2004.
Contents lists available at SciVerse ScienceDirect
Microchemical Journal journal homepage: www.elsevier.com/locate/microc
Evaluation of some wet digestions methods for reliable determination of total phosphorus in Australian soils Benjamin Webb, Samuel B. Adeloju ⁎ NanoScience and Sensor Technology Research Group, School of Applied Sciences and Engineering, Monash University, Gippsland Campus, Churchill, Victoria 3842, Australia
a r t i c l e
i n f o
Article history: Received 30 June 2012 Received in revised form 27 January 2013 Accepted 1 February 2013 Available online 9 February 2013 Keywords: Phosphorus Total P concentration Soil Wet digestion Acid digestion
a b s t r a c t Four common wet digestion methods have been evaluated for reliable determination of total phosphorus (P) concentrations in soil samples. Wet digestion of soil samples with nitric acid gave the highest recovery of total P concentrations with a percentage recovery of 90.1±0.9% (n=3). A lower percentage recovery of 87.0±1.4% (n=3) was achieved by wet digestion of soil samples with sulphuric acid. The use of acid mixture or acid–alkaline mixture for wet digestion of soil samples gave phosphorus recoveries of 82.4±1.9% (n=3) and 85.4±2.1% (n=3) with nitric acid–sulphuric acid mixture and nitric acid–potassium persulphate mixture, respectively. Substantial improvement in phosphorus recoveries with wet digestion of soil samples with sulphuric acid was achieved by further treatment of digested soil samples with sulphuric acid, resulting in a recovery of 92.8±1.0% (n=3), which was higher than possible with other acid and mixtures. The wet digestion of soil samples with sulphuric acid was also the only method which met reactivity and safety considerations. The successful utilisation of wet digestion with sulphuric acid for reliable determination of total P concentrations in a range of soil samples from some Australian dairy and beef rearing pastoral land is reported. © 2013 Elsevier B.V. All rights reserved.
1. Introduction Phosphorus (P) is an essential nutrient in natural ecosystems and it is also critically important to agricultural systems [1–6]. It is required for many agricultural processes including photosynthesis, nitrogen fixation, flowering, seeding and fruit maturation [3–6]. Soils in most parts of the world are low in P and deficient for optimal agricultural production [1,4–8]. This deficiency is often compensated for by applying various sources of P to soil, and thereby increasing the P status of the soil and its agricultural yield. Unfortunately, this mode of addition of P to soils has also had severe and detrimental effects on several waterways around the world, leading to an increased rate of eutrophication [9–12] and common presence of algal blooms which have been associated with various problems in fish, livestock and human health [4,12–14]. The presence of phosphorus (as phosphate) and nitrogen (as nitrate) has also been implicated in many cases with several reported outbreaks of blue-green algae which can have devastating effects on waterways. A very good indication of the soil P pools and management practices that may contribute to P enrichment of runoff and waterways can be obtained by conducting soil P measurements [5,6,12,15–17]. These measurements provide a basis for matching P inputs and agricultural crop demands [1,5,6,18]. Soil P measurements have been used to assess both environmental and agronomic impacts of P concentration extractable from soil [6,18–21]. ⁎ Corresponding author. Tel.: +61 3 9902 6450; fax: +61 3 9902 6738. E-mail address: [email protected] (S.B. Adeloju). 0026-265X/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.microc.2013.02.001
Total P concentration in soil is an important parameter which accounts for all forms of P within the soil. This parameter is often used to determine soil P status for phosphorus based fertiliser application and for estimation of P exports or agricultural yield [5,6,22]. It also provides a useful indication of the overall and potential nutrient supply of P, and has been used in relationship comparisons with other soil measurements. However, total P measurement is limited in that it does not differentiate between plant available P and non-available sources, such as organically bound from insoluble mineral P [23]. Nevertheless, it is still a very significant parameter for both environmental and agronomic considerations. The reliable determination of total P concentrations in soil is however not easy. It requires adequate and effective decomposition of the soil matrix to ensure complete release of P into solution prior to analytical measurement. A number of digestion methods have been proposed for the determination of total P concentrations in soil. In one method, Tan [24] used a fluoro-boric acid digestion which employed a specialised bomb digestion vessels and hydrofluoric acid. The original version of this method [25] also required specialised platinum crucibles and utensils for sodium hydroxide or sodium carbonate fusion [24]. Olsen and Sommers [26] also proposed a sodium carbonate fusion method, and this again required specialised platinum equipment. Another suggestion from Olsen and Sommers [26] involved wet digestion of soil with a perchloric acid. This method was not only complex, but had a range of safety issues and required a specialised perchloric acid fumehood. For these reasons, this digestion method is rarely used, except in specialised laboratories with adequate perchloric acid fumehood. Due to safety concern, some of the available soil method handbooks have
48
B. Webb, S.B. Adeloju / Microchemical Journal 111 (2013) 47–52
deliberately not included methods that employ perchloric acid [23]. However, it is important to note that safety consideration is not only limited to the use of perchloric acid, but also to the reactivity of some acids and/or mixtures to soil samples. In another study, Dick and Tabatabai [27] reported on the use of a sodium hypobromite/sodium hydroxide solution for total P determination in soil samples. The mixture was heated to dryness on a sand bath at 260 °C, followed by addition of formic acid and sulphuric acid. Bowman [28] also used ordered additions of sulphuric acid, hydrogen peroxide and hydrofluoric acid to decompose soil samples for total P determination. Both of these methods require considerable manipulation of the sample and are labour intensive compared to other methods [29]. The examples cited above highlight the current state of play with the determination of total P concentrations in soil. This is obviously not ideal for such a significant parameter which is highly important in assessing the inputs of P into farmland and waterways. In general, most of the currently available methods require specialised laboratory conditions or equipment, dangerous chemicals and intensive labour. To address this issue and ensure ease of attainment of reliable total P concentrations, there is a need for the development of a simpler soil digestion method that can be readily employed within standard laboratory conditions without safety concern, complex and laborious processes. This paper reports on a thorough evaluation and assessment of four wet digestion methods carried out to enable identification and/ or development of a simple and direct approach for pre-treatment of soil samples for reliable determination of total P concentrations. The wet digestion methods considered are those that are readily used for determination of other substances and that require the use of simple laboratory equipment, glassware and reagents [30,31]. The adequacy, efficiency and choice of the digestion methods were assessed on the basis of recovery efficiencies for total P concentration in soil samples, the ease of use, reproducibility and safety consideration. The application of the chosen method to the reliable determination of total P concentration in a wide range of soil samples from two intensive agricultural areas was also considered. 2. Materials and methods 2.1. Reagents and standard solutions All acids and reagents used were of analytical reagent grade solutions. All solutions were prepared and diluted with distilled deionised water (18 MΩ cm, Millipore, MA, USA). Nitric acid (0.01 M) used for adjusting the pH was prepared by diluting concentrated nitric acid with distilled deionised water. Stock phosphate solution (1000 mg P/L) was prepared by dissolving 4.393 g of potassium di-hydrogen orthophosphate in 1 L of distilled deionised water. A standard phosphate solution (100 mg P/L) was prepared by diluting an aliquot of this solution further with distilled deionised water. 2.2. Instrumentation All phosphate analyses were carried out by using an adapted method (SmartChem 140 Method 420-3651) modified for use with soil samples on the Westco SmartChem 140 automated wet chemistry discrete analyser (Westco Scientific Instruments, Inc., Brookfield, CT, USA). This method utilises an antimony–phospho-molybdate complex formed through the reaction of ammonium molybdate, antimony potassium tartate and dilute phosphorus solutions in an acid medium. Ascorbic acid is added to reduce the complex to produce a blue coloured complex measured at 880 nm. The resulting absorbance increased proportionally to P concentration in solution. The normal sample and reaction diluant in the method was deionised water, but was changed in this study to 0.01 M nitric acid to match the acidity of diluted sample digest.
2.3. Glassware All glassware and other containers were washed, soaked in a 2 M nitric acid for at least 7 days, rinsed three times with deionised water, soaked in deionised water and finally soaked in 0.1 M hydrochloric acid (HCl) until ready for use. 2.4. Heating sources Standard laboratory hotplates with aluminium surfaces were used for heating samples. The maximum temperature setting was used for each of the methods or as safety permitted. The temperatures of the extracts were recorded and their significance for sample digestion is discussed later. 2.5. Sample collection and preparation Soil samples were collected from an irrigation bay (width 30 m, length 300 m) at the Macalister Research Farm (38°00′S 146°54′E), a dairy farm situated in the Macalister Irrigation District of south-east Victoria (Australia). The soil was a natric grey Sodosol [32] and carried pastures that contained perennial ryegrass (Lolium perenne), white clover (Trifolium repens) and assorted invasive species including dock (Rumex spp.) and distichum (Paspalum paspaloides) [11,33]. 200 samples of 20 mm cores and 30 samples of 100 mm cores were collected from the sampling sites using a grid pattern. The soil cores were bulked for each depth to provide a composite sample for the sampling location. After collection, the soil was stored (4 °C) in polyurethane bag and transported to the laboratory. The bulked soil cores were air dried (40 °C), ground and passed through a 2 mm sieve. Samples were then stored in polyethylene containers at ca. 20 °C prior to analysis. Other soil samples were collected from selected agricultural sites within south-east Victoria, from 2 areas known as Maffra and Warragul. These areas were selected as geographically close agricultural areas (approximately 120 km range) with varying agricultural management practices, particularly irrigation application. A total of 14 sites were sampled, seven from each of the two areas. Soils were classified using an Australian soil classification system. The 7 sites in Warragul were of three different soil classifications, as indicated later in the results. The 7 sites in Maffra were also of 3 different soil classifications, as indicated later in the results. 2.6. Soil moisture content The moisture content was determined by heating the soil samples in a drying oven at 105 °C, cooled in desiccators and weighed repeatedly until a constant mass was obtained. The total P concentrations for each sample replicate were corrected for soil moisture content. 2.7. Digestion methods Four wet digestion methods were investigated for reliable determination of total P concentrations in soil samples. These methods were adapted from those reported previously by Adeloju et al. [30,31] for trace metal analyses. Digestion of each sample was carried out in triplicate. The specific soil digestion procedures for each of the wet digestion methods are described as follows: Method A: Nitric acid (HNO3) digestion: 0.3 g of soil sample was accurately weighed into a 100 mL Erlenmeyer flask. The flask containing sample was then transferred to a fumehood, where 10 mL of HNO3 (70%) was added and a glass funnel was inserted into the neck of the flask. The mixture was heated on a hotplate to approximately 125 °C where nitrogen oxide fumes were evolved and the volume of the mixture was reduced to approximately 2 mL. The flask was then
B. Webb, S.B. Adeloju / Microchemical Journal 111 (2013) 47–52
allowed to cool for at least 3 min. This process was repeated twice with another addition of 10 mL HNO3 each time. With the final nitric acid addition, heating was continued at 125 °C until no further nitrogen oxide fumes were given off. The funnel was then rinsed with a small volume of deionised water into the flask and the contents were allowed to cool to room temperature. The contents were transferred to a 25 mL volumetric flask and made up to the mark with deionised water. Samples were allowed to settle overnight, and an aliquot was taken by pipetting for analysis or storage. Samples were then stored at −20 °C in acid washed polyethylene containers until ready for processing. Method B: Sulphuric acid (H2SO4) digestion: 0.3 g of soil sample was accurately weighed into a 100 mL Erlenmeyer flask. The flask containing sample was then transferred to a fumehood, where 10 mL of H2SO4 (98%) was added and a glass funnel was inserted into the neck of the flask. The mixture was heated on a hotplate to approximately 160 °C until sulphite fumes were evolved and the volume of the mixture was reduced to approximately 2 mL. The flask was then allowed to cool for at least 3 min. This process was repeated 4 more times with 1 mL H2SO4 addition each time. With the final addition, heating was continued at 160 °C until no further sulphite mist was given off. The funnel was then rinsed with a small volume of deionised water into the flask and the contents were allowed to cool to room temperature. The contents were transferred to a 25 mL volumetric flask and, the digestion flask was rinsed carefully by using deionised water, and the content of the volumetric flask was mixed to avoid an acid–water inversion layer. The content of the flask was allowed to cool before being made up to the mark with deionised water. Samples were allowed to settle overnight, and an aliquot was taken by pipetting for analysis or storage. Samples were then stored at −20 °C in acid washed polyethylene containers until ready for processing. Method C: Mixture of nitric and sulphuric acids (HNO3–H2SO4) digestion: 0.3 g of soil sample was accurately weighed into a 100 mL Erlenmeyer flask. The flask containing sample was then transferred to a fumehood, where 10 mL of HNO3 (70%) and 1 mL H2SO4 (98%) were added and a glass funnel was inserted into the neck of the flask. The mixture was heated on a hotplate to approximately 130 °C where nitrogen oxide fumes were evolved and the volume of the mixture was reduced to approximately 2 mL. The flask was then allowed to cool for at least 3 min. This process was repeated 3 times with 10 mL HNO3 addition each time. With the final addition, heating was continued at 130 °C until no further nitrogen oxide fumes were given off. The funnel was then rinsed with a small volume of deionised water into the flask and the contents were allowed to cool to room temperature. The contents of the Erlenmeyer flask were transferred into a 25 mL volumetric flask and made up to the mark with deionised water. Samples were allowed to settle overnight, and an aliquot was taken by pipetting for analysis or storage. Samples were then stored at −20 °C in acid washed polyethylene containers until ready for processing. Method D: Mixture of nitric acid and potassium persulphate (HNO3– K2S2O8) digestion: 0.3 g of soil sample was accurately weighed into a 100 mL Erlenmeyer flask. The flask containing sample was then transferred to a fumehood, where 10 mL of concentrated HNO3 (70%) and 4 mL K2S2O8 (10% m/V) were added and a glass funnel was inserted into the neck of the flask. The mixture was heated on a hotplate to approximately 120 °C where nitrogen oxide fumes were evolved and the volume of the mixture was reduced to approximately 2 mL. The flask was then allowed to cool for at least
49
3 min. This process was repeated with 10 mL HNO3 addition each time. With the final addition, heating was continued at 120 °C until no further nitrogen oxide fumes were given off. The funnel was then rinsed with a small volume of deionised water into the flask and the contents were allowed to cool to room temperature. The contents of the Erlenmeyer flask were transferred into a 25 mL volumetric flask and made up to the mark with deionised water. Samples were allowed to settle overnight, and an aliquot was taken by pipetting for analysis or storage. Samples were stored at −20 °C in acid washed polyethylene containers until ready for processing. 2.8. Recovery spikes For each digestion method, a series of standard spikes were employed to assess recovery efficiency. The spike additions employed were: 0 (blank), 500, 1000 and 1500 mg P/kg. The spiked amount was added to individual soil samples prior to the addition of the digestion acid or mixture. Each sample spikes were also repeated in triplicate. 2.9. Sample preparation and calibration A 2 mL aliquot of each sample digest was diluted in a 25 mL volumetric flask with 0.01 M HNO3 solution prior to analysis. The resultant solution was then analysed for P concentration by using the SmartChem 140 discrete analyser. The results of the replicate samples were then combined and analysed statistically. Calibration was carried out using a six point calibration with 0.0, 0.5, 1.0, 2.0, 3.1 and 5.0 mg P/L. The SmartChem 140 discrete analyser automated system prepared a linear calibration plot through the dilution of a 100 mg P/L stock solution. The R 2 values for the calibration plots ranged from 0.9841 to 0.9922. 2.10. Method/reagent blanks Blank digestions of reagents were prepared for each of the 4 wet digestion methods. The same digestion procedure outlined for each digestion method was followed for each blank without the addition of soil sample. Final concentrations were corrected for the blank solution P concentrations for each method. The average blank contribution of the digestion solutions from each method were: method A (0.19 ± 0.04%), method B (0.20 ± 0.04%), method C (0.15 ± 0.02%) and method D (0.20 ± 0.05%). Evidently, there were no significant differences between the blank corrections for the four wet digestion methods. 2.11. Working environment All sample preparation and analyses were conducted in controlled laboratory conditions at 22.0 ± 0.5 °C. All digestions were undertaken in standard laboratory fume hoods. 3. Results and discussion 3.1. Recovery study As a first step in assessing the effectiveness of the four chosen wet digestion methods for reliable determination of total P concentrations in soil, their recovery efficiencies were compared and evaluated. As the two broad type of soil samples collected from the sampling sites in this study are “clay loam” and “sandy loam”, we chose to use a clay loam for the comparison because the soil samples collected from 13 of the 14 sampling sites were clay loam of different colour variations. It was not expected that the colour variation will alter
50
B. Webb, S.B. Adeloju / Microchemical Journal 111 (2013) 47–52
the ability to acid digest the clay loam samples. The chosen digestion method will be subsequently applied to the sandy loam sample. Soil samples digested with only concentrated HNO3 (method A) gave the highest percentage recovery for total P concentration, averaging 90.1± 0.9%. On the other hand, soil samples digested with only concentrated sulphuric acid (method B) gave a slightly lower recovery of 87.0± 1.4%. Contrary to expectation, the use of a mixture of concentrated nitric and sulphuric acids (method C) and a mixture of concentrated nitric acid and potassium persulphate (method D) gave even lower recoveries of 82.4± 1.9% and 85.4 ±2.1%, respectively. In particular, the lowest recovery efficiency achieved with method C was somewhat surprising. The expectation was that the increased strength of the two strong acid mixtures would enable better decomposition of the soil matrix and, at least, produce similar or better percentage recovery of total P concentration compared to that obtained with the use of only concentrated nitric acid. To clarify the above unusual trend of the recovery efficiencies obtained for the four wet digestion methods, further investigation was conducted with unspiked soil samples. Interestingly, when these four digestion methods were applied to the determination of total P concentrations in unspiked soil samples, the data in Table 1 revealed that the phosphate concentrations found increased in the following order: Method B >
Method C > Method A > Method D:
This trend suggests that the use or inclusion of H2SO4 may be beneficial for more effective release of organically bound P in soil. By taking the total P concentration obtained for the soil sample with method B as optimum (100%), the estimated recoveries of P in the unspiked soil samples for the three other methods were 93.1% (method A), 95.9% (method C) and 76.1% (method D). Based on this assumption, a notable but significant observation is that the application of methods B and C to spiked soil samples (with 87.0 ± 1.4% and 82.4 ±1.9% recoveries for methods B and C, respectively) results in a net decrease of 13–14% in each case. As the presence of H2SO4 is a common factor in both digestion methods, this observation seems to suggest that the lower recoveries obtained in the spiked soil samples may be due to an impact of this acid on the inorganic P spike. A possible link that can be deduced from this observation is that the much higher digestion temperature achieved with H2SO4 only (method B) or as a mixture (method C) may result in losses of the inorganic P spikes during soil digestion. As can be clearly seen in Table 1, the digestion methods which include
Table 1 Recovery efficiency of wet digestion methods for total P concentrations in a grey clay loam Dermosol. Sample description Method Method Method Method Method Method Method Method Method Method Method Method Method Method Method Method a b c d
A unspikeda A spike 1a A spike 2a A spike 3a B unspikedb B spike 1b B spike 2b B spike 3b C unspikedc C spike 1c C spike 2c C spike 3c D unspikedd D spike 1d D spike 2d D spike 3d
Max digestion temp. °C
P spike conc. (mg P/kg)
Total P (mg P/kg)
Recovery efficiency (%)
125
0 500 1000 1500 0 500 1000 1500 0 500 1000 1500 0 500 1000 1500
1777 ± 56 2229 ± 113 2683 ± 133 3111 ± 59 1908 ± 86 2346 ± 62 2788 ± 68 3189 ± 55 1829 ± 62 2249 ± 76 2656 ± 147 3034 ± 89 1452 ± 23 1893 ± 36 2296 ± 68 2720 ± 44
– 90.5 90.7 89.0 – 87.5 88.0 85.4 – 84.0 82.7 80.3 – 88.1 84.4 84.5
160
130
120
Concentrated HNO3 (3 × 10 mL) additions only (n = 3). Concentrated H2SO4 (1 × 10 mL and 4 × 1 mL) additions only (n = 3). Concentrated HNO3 (4×10 mL) and concentrated H2SO4 (1×1 mL) additions (n=3). Concentrated HNO3 (3×10 mL) and 10% w/v K2S2O8 (1×4 mL) additions (n=3).
H2SO4 (methods B and C) achieved higher solution temperatures, up to 40 °C higher than with other two digestion methods. Evidently, this is why method B gave the highest phosphate concentrations in the unspiked soil samples. Interestingly, the achieved solution temperature with the four digestion methods increased in the following order: Method B > Method C > Method A > Method D which correlated well with the trend observed for the total P concentrations in the unspiked soil samples. Notwithstanding the observed differences in the recovery efficiencies, the total P concentrations found in the unspiked soil samples within 0–20 mm depth by the four different digestion methods were comparatively close to those previously obtained at the same farm in 2001 (1528 mg P/kg) and 2004 (1432 mg P/kg) [10,11]. The higher concentrations found with methods A, B and C may be attributed to increasing P inputs into the sampled paddock between 2004 and 2007. 3.2. Choice of digestion method Apart from the recovery efficiencies achieved with the four wet digestion methods for total P concentrations in spiked and unspiked soil samples, proper selection of a digestion method must also consider relevant and/or associated safety issues. One of the important considerations in proposing a digestion method that can be readily employed in most laboratories for the determination of total P concentration in soils is the degree and extent of reactivity caused by the digestion medium or mixture during the digestion process. Highly reactivity digestion process can result in unexpected losses of samples and, in extreme cases, can result in accidental acid splash, corrosive reactions or other undesirable consequences. It is, therefore, highly essential that the chosen digestion method for soil samples is moderately reactive, while still ensuring complete recovery of total P concentrations in soils. The three wet digestions methods (methods A, C and D) which employed HNO3, either alone or as a mixture, were found to be very reactive, causing a sudden release of pressure when heated to the maximum temperatures. A pressure build-up below the fine particulate of the soil samples occur during digestion and led to the sudden pressure release. This can pose a serious safety concern as the digest can escape in some cases from the heated flask. This was particularly a problem with method A which used only HNO3 for soil digestion. In comparison, the pressure release and associated popping effect was somewhat subdued in method C and this appears to be due to the inclusion of H2SO4 in the mixture at the initial stage. However, more popping effect was observed with the required subsequent additions of HNO3 during digestion with this method. As a precaution, irrespective of the HNO3 based digestion method, care must be exercised and the heating temperature may need to be reduced to minimise potential hazard, but this will lead to a considerable extension of the digestion time. In contrast, no pressure build-up or popping effect was observed for sample digestion with method B, as the boiling temperature of H2SO4 was much higher. On the basis of recovery efficiency achieved for total P concentrations in unspiked soil samples and safety considerations, method B was chosen as the best of the four wet digestion methods for further investigation. Even with spiked soil samples, the recoveries obtained by this method was only slightly lower than those obtained with method A and were higher than those of methods C and D. Possible improvement in the recovery efficiency achieved with method B for determination of total P concentrations was further investigated by examining the effect of additional treatment with H2SO4. 3.3. Extended H2SO4 digestion (method B +) The possibility of improving the recovery efficiency achieved for spiked soil sample with method B was investigated by subsequent treatment of the initial sample digest with several 1 mL H2SO4 additions.
B. Webb, S.B. Adeloju / Microchemical Journal 111 (2013) 47–52
Table 2 Influence of increasing addition of H2SO4 on recovery efficiency of total P in a grey sandy loam Dermosol. Sample description Method Method Method Method Method Method Method Method Method Method Method Method a
B5 B5 B5 B5 B6 B6 B6 B6 B7 B7 B7 B7
unspiked spike 1 spike 2 spike 3 unspiked spike 1 spike 2 spike 3 unspiked spike 1 spike 2 spike 3
Additionsa (1 mL H2SO4)
Spike conc. (mg P/kg)
Total P (mg P/kg)
Recovery efficiency (%)
5 5 5 5 6 6 6 6 7 7 7 7
0 500 1000 1500 0 500 1000 1500 0 500 1000 1500
1083 ± 15 1539 ± 32 2000 ± 37 2449 ± 12 1092 ± 18 1549 ± 38 2018 ± 16 2468 ± 20 1123 ± 13 1585 ± 46 2062 ± 38 2505 ± 19
– 91.3 91.8 91.1 – 91.4 92.6 91.7 – 92.3 93.9 92.1
Please note: 0–100 mm depth samples (n = 3) were used.
96
Recovery Efficiency (%)
Method B, as described in materials and methods, involved digestion of soil sample in 10 mL concentrated H2SO4 followed by four 1 mL additions. Following exactly the same procedure, the effect of increasing the 1 mL H2SO4 additions beyond four on the recovery efficiency for total P concentration from spiked samples was of particular interest. The new methods (B5, B6, and B7 listed in Table 2 with 5, 6 and 7 additions of 1 mL H2SO4, respectively) were carried out on soil samples collected from 0 to 100 mm depth instead of the 0–20 mm depth used for comparison of the four wet digestion methods. For this reason, it was obvious that the total P concentrations obtained for these unspiked soil samples were lower than those reported in Table 1. The data in Table 2 show that the total P concentrations in the unspiked soil samples increased slightly with each 1 mL H2SO4 addition and treatment, indicating small improvement in the release of organically bound P. Also the percentage recoveries obtained for the spiked soil samples improved slightly with each acid addition. Evidently, the average spike recovery obtained increased to 91.4 ± 0.4% for method B5, to 91.9 ± 0.6% for method B 6, and further to 92.8 ± 1.0% for method B7. Fig. 1 shows the effect of increasing addition of H2SO4 on the average recoveries of total P concentration in the spiked soil samples. These results demonstrate that a net improvement in recovery efficiency of 4– 6% was achieved by further treatment of sample digest with 1–3 additions of 1 mL H2SO4. Furthermore, it is worth noting that the average spike recoveries obtained with methods B5, B 6, and B7 were all higher than that of method A. Fig. 1 also shows that the recovery of P with this approach increased considerably between the fourth and fifth 1 mL additions and then levelled out with the sixth and seventh additions. This observation clearly suggests that further acid additions will not lead to more improvement in recovery and was therefore not considered. These results also suggest that the lower P recoveries obtained earlier for the spiked soil samples collected from 0 to 20 mm depth may be due to other soil composition. The 0–20 mm soil is likely to contain higher amounts of organic material than those from 0 to 100 mm depth. This could mean that the soil samples collected from 0 to 20 mm depth were more difficult to digest. Interestingly, recovery study of spike 0–100 mm depth soil samples by the original method B digestion method gave an average recovery comparable to the original results obtained with the 0–20 mm depth samples (87.0% ± 1.4% for 0–20 mm depth versus 87.2% ± 0.9% for 0– 100 mm depth). However, it is recommended, based on the results obtained in this section that method B should be employed with at least five 1 mL H2SO4 additions. Table 3 shows that this approach was successfully employed for the determination of total P concentration in soil samples collected from two intensive agricultural areas. Evidently, the total P concentrations in these soil samples are influenced by factors such as the soil characteristics, quantity, frequency and length of P fertiliser application.
51
94 92 90 88 86 84 3
4
5
6
7
Number of 1 mL H2SO4 Additions Fig. 1. Average spike recovery (%) versus increasing number of 1 mL H2SO4 additions for wet digestion of grey sandy loam Dermosol with method B. n = 3.
4. Conclusions The careful evaluation of the four wet digestion methods for reliable determination of total phosphorus (P) concentrations in soil samples has demonstrated that the criteria for selecting an appropriate method must include safety consideration, in addition to the achievable recovery efficiency. Wet digestion with H2SO4 was found to be less reactive, most effective and safe for total P determination in soil. Further treatment of pre-digested soil samples with H2SO4 led to substantial improvement in phosphorus recovery, achieving a percentage recovery of 92.8 ±1.0%. This digestion method was successfully utilised for the reliable determination of total phosphorus concentrations in various soil samples from dairy and beef rearing pastoral areas.
Acknowledgements One of the authors (Ben Webb) wishes to acknowledge the provision of an Australian Postgraduate Research Scholarship through Monash Research Graduate School (MRGS) and other support provided by the School of Applied Sciences and Engineering. The authors also acknowledge the research funding provided for this project by Dairy Australia.
Table 3 Total phosphorus concentrations in soil samples collected from intensive agricultural areas in Warragul and Maffra (S-E Victoria, Australia). Sample site
Total P (mg P/kg)
Soil characteristica
Period post fert. appln. (day)
Last fert. appln. (mg P/ha)
W1 W2 W3 W4 W5 W6 W7 M1 M2 M3 M4 M5 M6 M7
1838 1941 3009 2680 1779 2062 1103 1936 2381 1447 1470 1775 2000 1743
GCLD GCLD RCLF RCLF RCLF GCLD GSLD GRCL GRCL GRCL GRCL GYCL GCLS GCLS
365 365 365 365 45b 20 730+ 365 45 1000+ 730+ 60 130 130
24.8 24.8 24.8 24.8 3.7b 14.8 n/a 11.8 14.5 n/a n/a 13.2 7.7 7.7
n/a = not available. a Soil description; GCLD=grey clay loam Dermosol, RCLF=red clay loam Ferrosol, GSLD=grey sandy loam Dermosol, GRCLS=grey/red clay loam Sodosol, GRCLS=grey/ yellow clay loam Sodosol, GCLS=grey clay loam Sodosol. b Site W5 has only received a single fertiliser application (3.7 kg P/ha) in three years prior to sampling.
52
B. Webb, S.B. Adeloju / Microchemical Journal 111 (2013) 47–52
References [1] P.W. Moody, M.D.A. Bolland, in: K.I. Peverill, L.A. Sparrow, D.J. Reuter (Eds.), Soils Analysis: An Interpretation Manual, CSIRO Publishing, Collingwood, Australia, 1999, p. 187. [2] R.W. McDowell, B.J.F. Biggs, A.N. Sharpley, L. Nguyen, Connecting phosphorus loss from agricultural landscapes to surface water quality, Chem. Ecol. 20 (2004) 1–40. [3] N.C. Brady, R.R. Weil, The Nature and Properties of Soils, 14th Revised ed. Pearson Prentice Hall, New Jersey, USA, 2008. [4] D. Cordell, J.O. Drangert, S. White, The story of phosphorus: global food security and food for thought, Global Environ. Change 19 (2009) 292–305. [5] J. Shen, L. Yuan, J. Zhang, H. Li, Z. Bai, X. Chen, W. Zhang, F. Zhang, Phosphorus dynamics: from soil to plant, Plant Physiol. 156 (2011) 997–1005. [6] K.W. King, J.C. Balogh, S.G. Agrawal, C.J. Tritabaugh, J.A. Ryan, Phosphorus concentration and loading reductions following changes in fertilizer application and formulation on managed turf, J. Environ. Monit. 14 (2012) 2929–2938. [7] In: D.J. Connor, D.F. Smith (Eds.), Agriculture in Victoria, Australian Institute of Agricultural Science, Parkville, Australia, 1987. [8] I.C.R. Holford, Soil phosphorus: its measurement and its uptake by plants, Aust. J. Soil Res. 35 (1997) 227–239. [9] R.B. Grayson, K.S. Tan, A. Western, Estimation of Sediment and Nutrient Loads into the Gippsland Lakes, CSIRO and University of Melbourne, Melbourne, Australia, 2001, p. 77. [10] D. Nash, B. Webb, M. Hannah, S. Adeloju, M. Toifl, K. Barlow, F. Robertson, F. Roddick, N. Porter, Changes in nitrogen and phosphorus concentrations in soil, soil water and surface run-off following grading of irrigation bays used for intensive grazing, Soil Use Manage. 23 (2007) 374–383. [11] D. Nash, M. Hannah, K. Barlow, F. Robertson, N. Mathers, C. Butler, J. Horton, A comparison of some surface soil phosphorus tests that could be used to assess P export potential, Aust. J. Soil Res. 45 (2007) 397–400. [12] In: E.K. Bünemann, A. Oberson, E. Frossard (Eds.), Phosphorus in Action: Biological Processes in Soil Phosphorus Cycling, 1st edition, Springer, 2011(XV). [13] A.M. Ebeling, L.R. Cooperband, L.G. Bundy, Phosphorus source effects on soil test phosphorus and forms of phosphorus in soil, Commun. Soil Sci. Plant Anal. 34 (2003) 1897–1917. [14] R. McDowell, A. Sharpley, P. Kleinman, Surface water quality and phosphorus applications, Encycl. Water Sci. 1 (2003) 961–964. [15] T.M. Aye, M.L. Nguyen, N.S. Bolan, M.J. Hedley, Phosphorus in soils of riparian and non-riparian wetland and buffer strips in the Waikato area, New Zealand, N.Z. J. Agric. Res. 49 (2006) 349–358. [16] A.N. Sharpley, S.J. Smith, Prediction of soluble phosphorus transport in agricultural runoff, J. Environ. Qual. 18 (1989) 313–316.
[17] A.N. Sharpley, R.W. Tillman, J.K. Syers, Use of laboratory extraction data to predict losses of dissolved inorganic phosphate in surface water and tile drainage, J. Environ. Qual. 6 (1977) 33–36. [18] N. Devau, E. Le Cadre, P. Hinsinger, F. Gérard, A mechanistic model for understanding root-induced chemical changes controlling phosphorus availability, Ann. Bot. (Lond) 105 (2010) 1183–1197. [19] A.M. Atia, A.P. Mallarino, Agronomic and environmental soil phosphorus testing in soils receiving liquid swine manure, Soil Sci. Soc. Am. J. 66 (2002) 1696–1705. [20] A.N. Sharpley, Soil phosphorus dynamics: agronomic and environmental impacts, Ecol. Eng. 5 (1995) 261–279. [21] G.M. Pierzynski, J.T. Sims, G.F. Vance, Soils and Environmental Quality, CRC Press, Boca Raton, Fl, USA, 2005. [22] D. Kara, C. Özsavasci, M. Alkan, Investigation of suitable digestion methods for the determination of total phosphorus in soils, Talanta 44 (1997) 2027–2032. [23] G.E. Rayment, F.R. Higginson, Australian Laboratory Handbook of Soil and Water Chemical Methods, Inkata Press, Melbourne, Australia, 1992. [24] K.H. Tan, Soil Sampling, Preparation and Analysis, 2nd ed. CRC Press, Boca Raton, Fl, USA, 2005. [25] B. Bernas, A new method for decomposition and comprehensive analysis of silicates by atomic absorption spectrometry, Anal. Chem. 39 (1968) 1632–1636. [26] S.R. Olsen, L.E. Sommers, in: A.L. Page, R.H. Miller, D.R. Keeney (Eds.), Methods of Soil Analysis, Part 2. Agronomy No. 9, American Society of Agronomy, Madison, WI, USA, 1982, pp. 403–430. [27] W.A. Dick, M.A. Tabatabai, Determination of orthophosphate in aqueous solutions containing labile organic and inorganic phosphorus compounds, J. Environ. Qual. 6 (1977) 82–85. [28] R. Bowman, A rapid method to determine total phosphorus in soils, Soil Sci. Soc. Am. J. 52 (1988) 1301. [29] I.P. O'Halloran, in: M.R. Carter (Ed.), Soil Sampling and Methods of Analysis, Lewis Publishers, Boca Raton, USA, 1993, pp. 213–230. [30] S.B. Adeloju, Comparison of some wet digestion and dry ashing methods for voltammetric trace element analysis, Analyst 114 (1989) 455–461. [31] S.B. Adeloju, A.M. Bond, M.H. Briggs, Critical evaluation of some wet digestion methods for the stripping voltammetric determination of selenium in biological materials, Anal. Chem. 56 (1984) 2397–2401. [32] R.F. Isbell, Australian Soil Classification, CSIRO Publishing, Melbourne, Australia, 1996. [33] B. Webb, D. Nash, M. Hannah, S. Adeloju, M. Toifl, F. Roddick, N. Porter, in: B. Singh (Ed.), Phosphorus Between Soil, Soil Water and Overland Flow for Established and Laser Graded, Border-check Irrigation Systems, SuperSoil 2004: Program and Abstracts for the 3rd Australian New Zealand Soils Conference, The Regional Institute Ltd, University of Sydney, Australia, 2004.