|
  
Evaluation of nutrient release rates from phosphate fertilisers in a laboratory leaching technique Chitralekha Ramakrishnan1 and Ken Perrott 2 1Research & Development Manager, Foremost Holdings Ltd, P.O. Box 12, Te Awamutu, New Zealand. Email: chitraleka@hotmail.com
2Scientist, Agresearch, Ruakura Research Centre, Hamilton, New Zealand Abstract
A laboratory leaching procedure was used to measure the release of P and Ca from a range of phosphate fertilisers [superphosphate, dicalcium phosphate, reactive rock phosphate (RPR)]. Periodic analyses of the leachate were used to follow the progress of P and Ca release over a 30 day period. The results showed that the water-soluble P decreased from 81% in super to 5% at 30% limestone content, and remained constant thereafter. Citric- and formic-acid soluble P decreased from 85% in super to 80% at 20% limestone content. These results indicate that at least 30% of limestone by weight should be added to super to ensure complete conversion from monocalcium (MCP) to dicalcium phosphate (DCP), and this result has been confirmed by XRD analyses. Although the water solubility of phosphate is decreased by this conversion, plant availability (as indicated by dilute acid solubility) is not likely to be affected. The release of P over time from super ranged from 75% at 1.5 hours to 85% after 30 days. This release rate decreased to 5% at 1.5 hours as the proportion of limestone increased to 30% and remained relatively constant thereafter. Cumulative Ca release was also less for the DCP products than it was from super. The cumulative P and Ca release of RPR was less than for all other fertilisers. The leaching rate shows that P and Ca released from RPR is greater than all other fertilisers after 15days and 27 days respectively. The study concludes that super releases P rapidly in a short time and the rate decreases over time. DCP releases P more slowly over a longer period than for fertilisers containing MCP. Keywords
Phosphate fertiliser, leaching, Citric soluble P, Ca/P molar ratio
Introduction
Superphosphate (Super) is the most commonly used phosphate fertiliser around the world and especially in New Zealand. The high level of super use recognises its importance on maintaining the production capacity of New Zealand’s agricultural sector. Recently a number of phosphate fertilisers have been introduced in to the market. The major reason for the continued introduction of modified P fertilisers is the increasing awareness of environmental impact of soluble fertilisers, and there is a trend towards using water-insoluble slow release P fertilisers.
Phosphate is usually the limiting nutrient for the formation of algal blooms in waterways, therefore identification and use of P fertilisers that can minimise P losses by leaching and runoff from pasture lands is beneficial for both increasing pasture P use efficiency and reducing P as an environment pollutant. In this view, some of the fertiliser companies in New Zealand have been producing and marketing di-calcium phosphate (DCP) fertiliser for the past one decade. Each company produces DCP using different raw materials following different methodology under the industry DCP standard of >70% citric soluble P in the product. However, there is only a limited knowledge available from previous research work on the nutrient release rate of DCP and its potential to reduce phosphate in waterways.
This paper attempts to investigate the nutrient release rates of DCP produced using different proportions of lime and Super and are compared with commercial fertilisers such as Super, lime and reactive phosphate rock (RPR) using a laboratory leaching technique.
Materials and Methods
The chemical composition of the various phosphate fertilisers used for the study is given in Table 1. A laboratory leaching procedure (Perrott and Kear 2000) using ion exchange resins to capture P and Ca released by the fertiliser products was used in this study. For detailed description of the procedure the above paper should be referred to.
With the fertilisers and lime, a defined particle size (2-4 mm) was used to limit particle size effects on leaching rate. Periodical samples were collected from the leaching tubes up to 30 days and analysed for Ca using atomic absorption spectrometry method and molybdate reactive orthoposphate-P by ascorbic acid/molybdate blue method (Murphy and Riley, 1962).
Manufacturing method of DCP: A range of dicalcium phosphate fertilisers were made in the laboratory by mixing limestone with super at limestone percentages by weight ranging from 0 to 50%. Once the reaction between the limestone and super was complete, the fertilisers were dried and passed through a nest of sieves (<0.5mm to >5mm). The 2 to 4mm particle size range was dominant in most of the samples and this was selected for analysis of total, citric- and formic-acid soluble and water soluble P and Ca. The agricultural lime used for the manufacture of DCP was sourced from McDonalds Lime, New Zealand (CaCO3 equivalent: >90%, Particle size distribution: <500 µ >50%, 2mm-500 µ 48%, > 2mm <2%).
Commercial DCP manufactured in the factory by Foremost Holdings Ltd, New Zealand and Gafsa RPR from Quinphos, New Zealand, were also used in the leaching experiment. Lime and super were mixed in proportion 70: 30 ratio for the treatment L/S 70.
Results
Characterisation of fertilisers
Chemical composition:
The fertiliser materials were tested at the lab, New Zealand, for pH (2:1 water:fertiliser), total P, citric-soluble P, formic-soluble P and water-soluble P, total Ca, citric-soluble Ca, formic-soluble Ca and water-soluble Ca using standard procedures.
Table 1: Chemical characterisation of investigated fertilisers
%
|
Total P
|
Formic P
|
Citric P
|
Water P
|
Total Ca
|
Formic Ca
|
Citric Ca
|
Water Ca
|
Super
|
9.2
|
7.8
|
8.0
|
7.5
|
22.4
|
11.7
|
11.9
|
9.0
|
95/05
|
8.9
|
3.8
|
7.2
|
0.3
|
23.5
|
12.7
|
12.4
|
6.6
|
90/10
|
8.1
|
4.0
|
6.5
|
0.3
|
24.3
|
13.2
|
13.1
|
5.7
|
85/15
|
7.8
|
4.2
|
6.3
|
0.3
|
24.9
|
16.0
|
15.7
|
3.9
|
80/20
|
7.0
|
5.1
|
5.5
|
0.3
|
25.6
|
16.8
|
16.2
|
3.7
|
75/25
|
6.4
|
5.6
|
5.0
|
0.4
|
27.1
|
19.2
|
18.5
|
3.0
|
70/30
|
6.8
|
5.0
|
5.9
|
0.5
|
28.8
|
24.1
|
22.6
|
2.2
|
65/35
|
6.3
|
5.5
|
5.3
|
1.4
|
28.6
|
24.8
|
23.1
|
2.1
|
60/40
|
5.5
|
6.2
|
4.4
|
2.2
|
28.6
|
22.7
|
22.3
|
2.4
|
55/45
|
5.2
|
6.5
|
4.1
|
4.2
|
28.7
|
23.9
|
22.9
|
2.2
|
50/50
|
4.6
|
7.2
|
3.7
|
5.6
|
29.6
|
25.6
|
24.7
|
2.2
|
DP 65
|
6.1
|
|
|
|
21.0
|
|
|
|
L/S 70
|
6.3
|
|
|
|
24.0
|
|
|
|
Gafsa
|
11.9
|
|
|
|
34.0
|
|
|
|
Figure 1: Percentage of P (a) and Ca (b) present in the fertiliser material as a percentage of Total P and Ca
Figure 1 shows the amount of P and Ca present as a percentage of total nutrients present in the fertiliser material. Super has the highest water soluble P, and the percentage decreased up to 75 % Super content and thereafter it is constant up to 50% super. However, the citric and formic acid soluble P remained at about 80% for all the materials. Unlike P, Ca gradually decreased from about 30% water soluble Ca to about 7% for 55% super and then increased to 30% for 50% lime-super mix. Citric and formic acid soluble Ca increased gradually from about 50% for super to 85% for 50% content.
Fertiliser pH and XRD analysis
Figure 2: Effect of fertiliser composition on (a) pH value and (b) composition
The measured pH values increase with lime content from 2.5 for super to 5.7 for a 70/30 super/lime mixture (Fig. 2a). With higher lime content the pH is approximately constant between 5.7 and 5.9.
X-ray diffraction analysis detected two phosphates (CaH2PO4, monocalcium phosphate and CaHPO4, dicalcium phosphate), calcite (CaCO3) and calcium sulphate (CaSO4, both hydrated and anhydrite) (Fig. 2b). The major component of super and the mixtures with lime up to 25% lime was CaSO4. With the higher lime contents the major components were CaSO4 and calcite, which were approximately equal. Monocalcium phosphate was highest in the super, declining with added lime to zero at 30% and higher lime contents. Dicalcium phosphate increased with lime content from zero for super to approximately 20% at 30% and higher lime contents. Calcite increased with lime content from zero for super up to 30 – 40% at 30% and greater lime contents.
Leaching of P and Ca
Cumulative P and Ca released
Figures 3a and 3b show the cumulative release curves of all the samples for P and Ca respectively. The cumulative release of P from the prepared materials was less than from super, particularly at short and intermediate times. Cumulative release of Ca was also less from the prepared materials than it was from super. The differences were greater for Ca than for P.
Figure 3: Leaching of (a) phosphate and (b) calcium by water
Most of the results are a result of the lime/super ratio of the samples, and Figures 4a and 4b give a clearer idea of the effect of composition on the release rates.
Figure 4: Leaching of (a) phosphate and calcium (b) at various times as a function of fertiliser composition
These figures show that the cumulative release of P decreased with decreasing super/lime ratio from a maximum for super and was approximately constant between compositions 70/30 and 50/50. At intermediate times the cumulative P release from materials 55/45 and 50/50 were slightly greater than from 70/30. At the longest time (30 days) the cumulative P release for the 70/30 to 50/50 materials was about 77%, less than for super (87%).
Release rate for P and Ca
The data were re-calculated to enable presentation of P and Ca release rates. Because of the large range of release rates found at different times and for different materials the presentation of meaningful results is not possible on single graphs. Thus, the data are presented in several graphs as a function of the super/lime ratio.
Figures 5a, b and c present the P release data at different times and Figs 6a and b do the same for Ca. At the shortest leaching time super had the greatest release rate for both P and Ca. The P release rate at 1.5 hr decline with lime content to a composition of about 75/25 and was constant between 75/25 and 50/50. The picture for Ca was similar. The Ca release rate at 1 day was similar for the compositions between 100% super and 85/15, generally declining for the compositions with greater lime content.
Figure 5: Release rate of P from the fertilisers at different times as a function of fertiliser composition
At longer times super had the lowest P and Ca release rates. The materials with the highest release rates at these times were those with a composition between 85/15 and 50/50 for P and 75/25 and 50/50 for Ca. At the longest time, lime had a greater Ca release rate than did the super/lime mixtures. There is a gradual decrease in the release of Ca over all time periods, with super recording the maximum. As the lime content increased, the release of Ca decreased up to 65 % super, beyond which the Ca content in the leachate increased.
Figure 6: Release rate of Ca from the fertilisers at different times as a function of fertiliser composition
Figure 7: Molar ratio (Ca/P) of Ca and P released at different times as a function of fertiliser composition
Molar ratio (Ca/P) of P release
The P and Ca release rate data were recalculated to show the ratio Ca/P (molar ratio) of the leachate at different times (Fig. 7).
At very short times (30 min) the Ca/P ratio is approximately 0.6 but increases with lime for lime contents of 20% and above to about 2.4 for the 50:50 material. At about 0.5 days the material being released from super has a very high Ca/P ratio (about 13) but decreases with lime content to about 2.9 for the 60/40 mixture. The picture is similar at about 2 days. Thereafter the Ca/P ratio of material being released from the super and low lime content samples declines to between 2.0 and 2.6.
The picture is different for the high lime materials where the Ca/P ratio of the released materials increases with time. The Ca/P ratio of materials of intermediate composition does not reach the high values seen for the other materials. The ratios settle to between 2 and 6.
Comparison of DCP with other phosphate fertilisers
Figures 8a and 8b show the cumulative leaching patterns of P and Ca respectively.
Figure 8: Cumulative leaching of (a) P and (b) Ca from a range of P fertilisers including DCP
The simple super/lime mixture (mixed immediately before leaching) had the same P release pattern as super. However the Ca release from this material was less than for super up to a leaching time of 25 days.
The cumulative release of P and Ca from the RPR was less than for all the other fertilisers. The P and Ca release patterns for the reacted super/lime materials were intermediate between those for RPR and for super.
Comparison of leaching rates
The super and fresh super/lime mixture had similar P release rates throughout the experiment (data not presented). These had the highest P release rates up to 0.5 days. The RPR had very low P release rates but after 15 days these were greater than for the other fertilisers. All the reacted super/lime materials had higher P release rates than either RPR or super between about 1 day and about 8 days. The P release rates at 13 days were roughly comparable to that for RPR.
Differences in the Ca release rate between the fertilisers were not as great as for P. However, the release rate from super was highest at the shortest times (30 and 90 min). At the longest times (5 days and more), the Ca release of the super rate was less than for the other fertilisers. The Ca release rate from RPR was greater than for the other fertilisers at the longest time (27 days).
Discussion
The reacted super/lime mixtures show the expected change of composition with super/lime ratio. The monocalcium phosphate (MCP) content was greatest for super and declined with increasing lime. The materials ranging from 70% super to 50% super contained no MCP. The dicalcium phosphate (DCP) content showed the opposite pattern, increasing from zero for super to about 20% for the 70% super material. The pH values of these materials reflect the changes, increasing from 2.5 for super to 5.7 for the 70% super material, which is consistent with reaction of MCP with lime.
The water insoluble nature of DCP as compared to water soluble MCP has been shown in Figure 1. Though there is low water soluble P present in DCP, the citric and formic acid soluble P forms, which determine plant availability, are unaffected (both MCP and DCP produce about 80% P). This shows that DCP produces an amount of plant available P equivalent to that produced by MCP.
The leaching data are interesting because they provide an indication of how the different materials may behave in the field, especially with respect to their ability to provide P for plant uptake. As we expected, P was leached most rapidly from the super. The proportion of fertiliser P leached at short times declined with lime content, reaching a minimum at about 70% super. This is consistent with the release of soluble MCP as found by XRD (Fig 2b).. The less soluble DCP released P at a slower rate than MCP, and at 30 days approximately 80% of P had been released from the materials with composition between 80 and 50% super, compared with 90% of the P in super.
Release of Ca during leaching with water shows some similarities to that for P. However, Ca was present in forms other than MCP and DCP. Consequently, the Ca leaching patterns also reflect dissolution of the different amounts of CaSO4 and free calcite present in the different materials.
The Ca/P molar ratios of the leachates at different times (Fig 7) provides an indication of the compounds influencing dissolution at different times. Thus, at short times (30 and 90 min) MCP (Ca/P = 0.5) was being leached from the super, 95% super and 80% super samples. With increasing lime, DCP (Ca/P = 1.0 for 75% super in 30 minutes) was being leached from the materials. However, with increasing time, the Ca/P ratio of the leachate increased markedly, particularly for the high super materials, because of dissolution of CaSO4. Leaching of Ca from calcite would have also contributed to the leachate Ca/P ratios. The high lime materials (55% and 50% super) show this with the increased Ca/P ratio at the longest times (Fig 7).
A closer examination of the rate of release of P at different times also shows the influence of amounts of MCP and DCP present in the different materials (Fig 5). The high P release rate at short times (1.5 hr) reflects the rapid release of P from MCP. On the other hand the lower P release rates at short times for materials with a composition between 80% and 50% super reflect the lower solubility of DCP. The P release rate of super and the high super materials rapidly declined (because of low amounts of remaining MCP) and at about 1 day the P release rates from all the material were similar (Fig 5). At 3 days and longer, P release from super and the high super materials (>90% super) was very low and much lower than the P release rates for 80% super – 50% super materials (Fig 5). These results are consistent with the ability of fertilisers containing DCP to release their P more slowly and hence over a longer period than for fertilisers containing MCP. The Ca release rates at different times (Fig 6) again are consistent with the release of Ca from MCP and CaSO4 at shorter times, from DCP at intermediate times, and from CaCO3 at longer times.
Comparison of P release from DCP with other P fertilisers
The results suggest that the DCP fertilisers would perform as slow-release fertilisers intermediate between super and RPR. Fig 8 indicates that whereas super releases its P over very short time periods, the DCP type fertilisers would release P over intermediate time periods. However, their performance compared with RPR would decrease over longer time periods. This simple description doesn’t take into account time scales in the field (laboratory leaching is extreme compared with the field) or the fact that P released from super would be adsorbed by soil and then act as a slow-release fertiliser. Nor are the effects of the chemical composition of the soil on dissolution rates taken into account. For example, the dissolution of RPR increases as soil pH decreases and exchangeable Ca decreases.
Conclusion
The reported results confirm the suggestion that DCP can be used as a slow-release fertiliser because DCP will release P more slowly and over a longer time than will MCP fertilisers such as super. While this has been used as an argument that DCP will perform better than super, it does not take into account interactions with the soil. For example, adsorption of dissolved P by the soil in effect produces a slow-release P fertiliser (P sorbed on soil). Future work should include glasshouse experiments with soils of different P retention properties to understand whether DCP behaves differently from super with respect to plant availability.
References
Murphy, J, Riley, P (1962). A modified single solution method for the determination of phosphate in natural waters. Analytica Chimica Acta 27, 31-36.
Perrott K, Kear MJ (2000). Laboratory comparison of nutrient release rates from fertilisers. Communications in Soil Science and Plant Analysis 31, 11-14. 
|
