Effect of application of molasses to paddy soil on the concentration of cadmium and arsenic in rice grain

Abstract

Molasses, a liquefied by-product of sugar production from sugarcane, is one of the most easily decomposable organic substances. We investigated the effect of the application of diluted molasses in paddy rice (Oryza sativa L.) fields on cadmium (Cd) and arsenic (As) absorption in rice grains as a measure of the shortage of irrigation water or rainfall. The application of diluted molasses on to the top of the soil surface at one week before or after heading led to a decrease in the redox potential of the soil. The Cd concentration in the rice grains was significantly decreased with the application of molasses (2000 kg ha⁻¹) at heading or one week after heading. However, early application of molasses (one week before heading) did not decrease the Cd concentration in the rice grains with downsizing of grain particle size. Meanwhile, As concentration in the rice grains did not change with early application at one week before heading, while it increased with application of molasses at heading or one week after heading. This method is useful especially for decreasing Cd concentration in rice grains in a region under lack of irrigation due to drought.

Keywords:

• arsenic
• cadmium
• molasses
• paddy field
• rice grain.

Introduction

In Japan, the maximum level of cadmium (Cd) in rice (Oryza sativa L.) grains should be lower than 0.4 mg Cd kg⁻¹ due to the revision of the Food Sanitation Act (MHLW 2010). Therefore, farmers are required to produce rice grains with Cd concentrations below the new regulation level. To reduce Cd concentration in rice grains, flooding management during the period from 3 weeks before heading until 3 weeks after heading time has been practically used in paddy field so far (MAFF and NIAES 2005). However, if the water supply to the paddy is limited by drought in this period, rice grains in which the Cd concentration exceeds the maximum level may be produced more often (MAFF 2004). So, an alternative technique for reducing Cd concentration in rice grains is required under conditions of less rain and shortage of irrigation. In addition to Cd, arsenic (As) is a toxic chemical pollutant, the risk of which for human health should be assessed. Rice is recognized as the main source of As uptake by humans (MAFF 2009). It is well known that there is a trade-off relationship between the absorption of Cd and As in paddy rice. For example, flooding management of paddy fields is effective for reducing the level of Cd in rice grains; however, anaerobic conditions in paddy soil lead to arsenic mobilization and, therefore, As uptake by rice tends to increase.

Molasses, an adhesive brown liquid with a specific gravity of 1.4, is a by-product in the industrial process of producing crude sugar from sugarcane (Saccharum officinarum) and sugar beets (Beta vulgaris ssp. vulgaris var. altissima). Recently, restoration methods for tomato fusarium crown disease using application of molasses to subsoil in greenhouses have been developed and extended widely in Japan (Shinmura 2003). The mechanism of the effectiveness is explained by the decrease of soil redox potential (Eh) caused by the application of molasses.

The main objective of the present study is to investigate the effect of application of diluted molasses in a rice paddy field, under the conditions of shortage of irrigation water, on the concentration of Cd and As in rice grain.

Materials and Methods

Field experiment

A paddy field polluted with a low level of Cd, located in Hokuriku district, Japan, was used for the present study. The plow layer of the experimental field contained about 0.5 mg Cd kg⁻¹ extracted with 0.1 mol L⁻¹ hydrochloric acid (HCl) (1:5 w/v) and 1.5 mg As kg⁻¹ extracted with 1 mol L⁻¹ HCl (1:5 w/v). The soil is classified as a Typic Hydraquents by US Soil Taxonomy (Soil Survey Staff 2010) and the texture consists of loam (clay, 14.8%; silt, 22.1%; sand, 63.1%). The concentration of total carbon in the plow layer is 46.3 mg kg⁻¹. The source of Cd pollution in the field was irrigation water or flooding residues containing Cd from a nearby abandoned zinc mine.

The experimental plots (6 m × 3.6 m) were located in a paddy field surrounded by a ridge, because the irrigation conditions should be strictly controlled. Three plots were prepared for each application. The control plots (no application of molasses) were located adjacent to each experimental plot so as to minimize the variation of Cd and As concentration in the soil.

The molasses used in the present study was obtained from the sugar beet industry and contained: solids (74.3%), sucrose (49.6%), total nitrogen (N) (0.4%), raffinose (2.6%), reducing sugar (0.4%), ash (8.4%), at a pH of 7.7 and a Stammer Color Value of 365 (analyzed by Nippon Beet Sugar Manufacturing Co., Ltd).

We investigated the amount and period of the application method of diluted molasses onto the surface of soil in the rice growing period for reducing the Cd concentration in rice grains. Rice seedlings of the variety Koshihikari (Oryza sativa L.) were transplanted to the field on 13 May 2009 and harvested on 25 September 2009. The rice was grown under flooded conditions for five weeks after transplanting, followed by drainage for the next three weeks, followed by intermittent irrigation to 1 August 2009. A basal fertilizer was applied at rates of 25, 11 and 21 kg ha⁻¹ for N, phosphorus (P) and potassium (K), respectively [25 kg ha⁻¹ for phosphorus pentoxide (P₂O₅) and potassium (K₂O) respectively]. Ammonium sulfate was also supplied to the soil surface at the rate of 125 kg ha⁻¹ (25 kg N ha⁻¹) as topdressing at the panicle formation stage.

2000 kg ha⁻¹ of the molasses was diluted 12-fold, then the diluted molasses was applied to the soil surface of the field (21.8 m²) on three different days (one week before heading time (−1W) on 6 August; at heading time (0W) on 13 August; one week after heading time (+1W) on 21 August) to investigate the most suitable application time for decreasing the Cd concentration in rice grain.

In addition, three different levels of molasses (1000 kg ha⁻¹, 2000 kg ha⁻¹, 4000 kg ha⁻¹ as undiluted solution) were applied to 21.8 m² of field at heading time. After application of the molasses, no water was irrigated from the canal except for rainfall.

Analysis of plants and soil

Eh in the field was measured using a Platinum (Pt) electrode installed at a depth of 5 cm in the field between 10 and 12 am by Eh Meter (HORIBA D-52, Japan). Four replicate measurements of Eh were performed in each field. The yield component was surveyed at harvest time to evaluate the effect of application of the molasses.

The chemical properties of the plow layer of the field and the Cd concentration in rice grain were determined as previously reported (Honma et al. 2009). The As concentration of the rice grains was determined by hydride generation atomic absorption spectrometry (wavelength: 193.7 nm) (Hitachi Z-5020, Japan), after the rice grains were digested and diluted. Four rice hills were selected from each plot for analysis at harvest time. The rice grains were removed from the shoot and air-dried to 15% moisture. The rice grains for analysis were sieved above 1.85 mm thickness. Additionally, rice grains were separated into six different particle thin sizes (1.8–1.85 mm, 1.85–1.9 mm, 1.9–2.0 mm, 2.0–2.1 mm, 2.1–2.2 mm, and greater than 2.2 mm) and the Cd and As concentrations of each sample were determined.

Statistical analysis was performed using Statcel 2 software (Masui 2007). Multiple comparisons between treatments were made by Tukey-Kramer test. Statistical significance of Cd and As concentrations in rice grain was determined using t-test.

Results

Growth and yield of paddy rice

Table 1 shows the yield and yield component of rice treated with diluted molasses in different periods and amounts. The influence of molasses application to the winnowed paddy weight, shoot weight, grain-straw ratio, unpolished rice yield and percentage filled spikelet were not significant compared with the no-application plot. However, early application of molasses (−1W) brought a statistical decrease in the thousand-kernel weight of the rice grains.

Table 1. Yield and yield component of rice treated with diluted molasses at different periods and in different amounts^†

	Winnowed paddy weight (a) (kg/a)	Shoot weight (b) (kg/a)	Grain-straw ratio (a)/(b)	Unpolished rice yield (kg/a)	Percentage filled spikelets (%)	Thousand-kernel weight (g)
No-application	66.4 ± 3.2abc	56.6 ± 3.4a	1.17 ± 0.01a	52.8 ± 2.3abc	92.3 ± 1.7a	22.1 ± 0.0a
1 week before heading	74.8 ± 1.2ab	64.6 ± 3.2a	1.16 ± 0.04a	59.4 ± 0.8ab	92.6 ± 2.8a	21.6 ± 0.1b
At heading	63.3 ± 1.0c	51.2 ± 1.6a	1.24 ± 0.02a	50.5 ± 0.9c	92.8 ± 1.0a	22.4 ± 0.1a
1 week after heading	66.2 ± 0.9bc	53.2 ± 1.6a	1.25 ± 0.02a	52.6 ± 0.8ac	91.1 ± 1.2a	22.4 ± 0.1a
1000 kg ha^−1‡	75.4 ± 1.7a	64.6 ± 2.0a	1.17 ± 0.01a	60.4 ± 1.1ab	90.5 ± 2.8a	22.1 ± 0.0a
2000 kg ha^−1‡	63.3 ± 1.0c	51.2 ± 1.6a	1.24 ± 0.02a	50.5 ± 0.9c	92.8 ± 1.0a	22.4 ± 0.1a
4000 kg ha^−1‡	74.9 ± 2.1a	60.8 ± 2.5a	1.23 ± 0.02a	58.6 ± 2.4abc	90.9 ± 2.4a	22.2 ± 0.1a
^†Rice grain above 1.85 mm with 15 % moisture.
^‡Application at heading. Values represent means ± standard deviation. (n = 3). Means in the same column that are followed by the same letter are not significantly different (P < 0.05).

Changes of soil Eh

Table 2 shows the changes of soil Eh during treatment period. The Eh was not different between the control plot and the treatment plot on 4 August before the molasses treatment. On 11 August, the soil Eh declined only in the −1W (one week before heading) molasses application, although the Eh of the other plots were slightly increased. The column “(B)–(A)” shows the changes of Eh value from 4 August to 11 August, indicating the statistical difference between molasses treatment at −1W and the other treatment plots. A similar trend was observed in the +1W plot, when the changes from 13 August to 21 August were calculated in the “(D)–(C)” column. However, no reduction of Eh by molasses treatment was observed in the 0W plot where molasses application was done on 13 August. No difference between before and after treatment showed in the “(C)–(B)” column. Meanwhile, a difference in Eh between the 4000 kg ha⁻¹ plot and the control plot was observed after 0W molasses treatment. A statistically significant difference was shown in the “(C)–(B)” and “(D)–(C)” columns. So, the reductions of Eh caused by the application of molasses were observed to be significant except for applying molasses under the condition of drastic change of soil Eh. Although the molasses application decreased Eh, Eh was not decreased below −150 mV in any plot, which is a standard level of Eh for insolubilizing Cd in soil solution, mainly in the form of cadmium sulfide.

Table 2. Changes of redox potential (Eh) during treatment period

	Eh(mV) on measurement date				Difference across measurement date
	4 Aug^(A)	11, Aug^(B)	18 Aug^(C)	24 Aug^(D)	(B)–(A)	(C)–(B)	(D)–(C)
No-application	−108.8 ± 65.1a^†	−38.8 ± 20.6ab	437.8 ± 43.7a	577.5 ± 65.1ab	70.0 ± 81.4a	476.5 ± 52.4a	139.8 ± 46.2a
1 week before heading (app. on 6 Aug)	−7.5 ± 59.9a	−108.0 ± 77.0b	397.0 ± 61.8a	447.0 ± 90.2b	−100.5 ± 35.6b	505.0 ± 75.0a	50.0 ± 54.5ab
At heading (app. on 13 Aug)	−43.3 ± 29.5a	9.5 ± 26.5a	453.0 ± 67.2a	588.3 ± 80.2a	52.8 ± 15.1a	443.5 ± 49.3a	135.3 ± 47.2a
1 week after heading (app. on 21 Aug)	−54.0 ± 32.2a	3.5 ± 19.5a	496.0 ± 36.0a	462.3 ± 57.4ab	57.5 ± 48.4a	492.5 ± 54.3a	−33.7 ± 27.0b
1000 kg ha^−1‡	−12.0 ± 76.7a	43.0 ± 38.9a	471.5 ± 46.9a	525.0 ± 23.0ab	55.0 ± 48.1b	428.5 ± 19.9ab	53.5 ± 35.1ab
2000 kg ha^−1‡	−43.3 ± 29.5a	9.5 ± 26.5ab	453.0 ± 67.2a	588.3 ± 80.2a	52.8 ± 15.1b	443.5 ± 49.3ab	135.3 ± 47.2a
4000 kg ha^−1‡	−120.3 ± 38.5a	66.0 ± 32.4a	412.8 ± 73.9a	444.3 ± 72.0b	186.3 ± 43.1a	346.8 ± 56.3b	31.5 ± 40.3b
^†Values represent means ± standard deviation (n = 4). Means in the same column that are followed by the same letter are not significantly different (p < 0.05).
^‡Application on 13 Aug.

Concentration of Cd and As in rice grains

Table 3 shows the effect of the application of molasses on Cd and As concentrations in rice grains. No significant difference of Cd and As concentrations in rice grains was observed by the application of molasses at one week before heading. However, the Cd concentration in rice grains was significantly decreased by 18–30% with the application of molasses at 0W or +1W compared with no molasses application. And the effect of decreasing Cd concentration in rice grains was observed with molasses application of as low as 1000 kg ha⁻¹. In particular, the application of 4000 kg ha⁻¹ brought about the highest decrease of Cd concentration, by 39%. In contrast, under the application of 1000–2000 kg ha⁻¹ of molasses, the As concentration in rice grains increased significantly. The application of 4000 kg ha⁻¹ of molasses tended to increase the As concentration, but this was not statistically significant. Thus, there was a trade-off relationship between the Cd and As concentrations in rice grains by the application of molasses.

Table 3. Effect of the application of diluted molasses on cadmium (Cd) and arsenic (As) concentrations in rice grain^†

	Cd (mg kg^−1)			As (mg kg^−1)
	App.(a)	No-app.(b)	(a)/(b)	App.(a)	No-app.(b)	(a)/(b)
1 week before heading	0.255 ± 0.03	0.244 ± 0.05	1.04	0.105 ± 0.02	0.105 ± 0.02	1.00
At heading	0.195 ± 0.03	0.237 ± 0.03	0.82*	0.110 ± 0.01	0.078 ± 0.01	1.41*
1 week after heading	0.164 ± 0.03	0.235 ± 0.03	0.70**	0.111 ± 0.01	0.095 ± 0.01	1.16*
1000 kg ha^−1‡	0.230 ± 0.05	0.298 ± 0.02	0.77*	0.119 ± 0.01	0.082 ± 0.01	1.46*
2000 kg ha^−1‡	0.195 ± 0.03	0.237 ± 0.03	0.82*	0.110 ± 0.01	0.078 ± 0.01	1.41*
4000 kg ha^−1‡	0.159 ± 0.03	0.260 ± 0.06	0.61*	0.118 ± 0.03	0.099 ± 0.03	1.19
^†Rice grain above 1.85 mm with 15% moisture. *(p < 0.05), **(p < 0.01).
^‡Application at heading. App., Application of molasses; No-app., Non-application of molasses (Control).

Relationship between Cd and As concentration and particle size of rice grain

Figure 1 shows the Cd and As concentrations in rice grains with different particle sizes. The Cd concentration in rice grains with thickness of less than 1.9 mm was higher than that of grains with thickness above 1.9 mm. In particular, the Cd concentration in rice grains from 1.8 mm to 1.85 mm was about 2.5 times higher than that in grains of above 2.2 mm. On the other hand, As concentration was almost the same irrespective of particle size. The Cd concentration in rice grains increased as the particle size of the rice grain decreased. These trends were observed both with or without molasses application.

Figure 1. Effect of application of molasses on cadmium (Cd) and arsenic (As) concentrations in rice grain among different particle sizes.

Discussion

It is well known that the availability of heavy metals in soil varies with fluctuations in Eh (Makino 2002). For example, when a paddy field is flooded and the soil is in a reductive condition, most of the Cd in the soil combines with sulfur (S) to form cadmium sulfide (CdS), which has a low solubility in water. However, under oxidative conditions, CdS is converted into cadmium sulfate (CdSO₄), which is soluble in water. Arao et al. (2009) reported that the absorption of Cd and As in rice grains depended on the changes of Eh in soil during the three weeks before and after heading. And they recognized the trade-off relationship between the Cd and As concentrations in rice grain. In this study, we observed the decrease of Cd and increase of As in rice grains with the application of molasses, followed by a decrease of Eh in soil (except when molasses was applied at heading time, when the changes in Eh were not different from those in the untreated plot). We considered that the drastic increase of Eh caused by soil dryness might mask the influence of the application of molasses. Thus, we considered that the application of molasses at heading to one week after led to reductions in Eh and to changes in the Cd and As concentrations in rice grains. However, the decrease of Eh by application of molasses was not enough for forming CdS (practically insoluble form in soil solution), of which Eh to change from CdSO₄ was −150 mV (Connell and Patrick 1968). Ito and Iimura (1975) observed the same phenomenon in a pot experiment. We considered that the reason why the soil Eh did not lower enough at 5 cm of soil depth to be because the percolation of molasses toward the subsoil was not enough due to topdressing of the molasses. Because of this result, it was suggested that the critical factor causing a decrease of Cd in rice grains was not only forming CdS in soil, but also other factor. So, we should search for the other factor which may be, for example, changes of nutrient absorption, or translocation of Cd in shoot, etc., caused by application of molasses.

The effect of decreasing Cd concentration in rice grains depended on the time of application of the molasses. No changes of Cd and As concentrations in rice grains were observed in the application at one week before heading. However, applying the molasses after heading caused a decrease of Cd and an increase of As concentration. One reason for this result may be the difference of particle sizes of rice grains. We confirmed the rice grain to be thin, when the molasses was applied at one week before heading, although no changes were observed in particle size after heading (Fig. 2). Ishiguro and Yamada (1997) recognized a statistically significant negative correlation between the weight of rice grains and the Cd concentration. They described that the Cd concentration of small rice grains, which accumulated the starch granules incompletely, had been increased relatively, because of the time gap to accumulate starch and Cd into the grains. So, the selection of rice grains by sieving is useful for decreasing the Cd concentration in rice grains before circulation (i.e., grains below 1.9 mm).

Figure 2. Effect of application of molasses on particle size distribution of rice grain. (Left: Effect of different applying periods; Right: Effect of applying different amounts). W, week.

Meanwhile, there was no difference in As concentration with different particle sizes. The As concentration in rice grains was increased by the application of molasses. Thus, it was suggested that As was accumulated in the grain at the same time as the starch, in contrast with Cd. These results lead to the possibility of decreasing Cd and As simultaneously because of the different stages of accumulating Cd and As into rice grains.

In conclusion, our study showed that application of diluted molasses after heading time is useful for decreasing the Cd concentration in rice grains, in the same way as flooding management. Further investigation is required to decrease the Cd and As concentrations simultaneously in rice grains.