WEl Hai-yan, CHEN Zhi-feng, XlNG Zhi-peng, ZHOU Lei, LlU Qiu-yuan, ZHANG Zhen-zhen, JlANG Yan,HU Ya-jie, ZHU Jin-yan, CUl Pei-yuan, DAl Qi-gen, ZHANG Hong-cheng

Jiangsu Key Laboratory of Crop Genetics and Physiology/Jiangsu Co-Innovation Center for Modern Production Technology of Grain Crops/Agricultural College, Yangzhou University, Yangzhou 225009, P.R.China

Abstract There is limited information about the influence of slow or controlled release fertilizer (S/CRF) on rice yield and quality. In this study, japonica rice cultivar Nanjing 9108 was used to study the effects of three different S/CRFs (polymer-coated urea(PCU), sulfur-coated urea (SCU), and urea formaldehyde (UF)) and two fertilization modes (both S/CRF and common urea(CU) as basal fertilizer, S/CRF as basal and CU as tillering fertilizer) on rice yield and quality. CU only was applied separately as control (CK). Results showed that, rice grain yield, chalky kernel rate, chalky area, overall chalkiness, and the content of gliadin, glutenin, and protein, all showed the trends of UF＞PCU＞SCU within the same fertilization mode, and showed the trends of S/CRF as basal and CU as tillering fertilizer＞both S/CRF and CU as basal fertilizer within the same type of S/CRF. In contrast, the contents of amylose, amylopectin, and starch, as well as taste value, and peak and hot viscosity showed trends of SCU＞PCU＞UF, and the trends of both S/CRF and CU as basal fertilizer＞S/CRF as basal and CU as tillering fertilizer. Among S/CRF treatments and fertilization modes, taste values of cooked rice were positively correlated with amylose, amylopectin, and starch contents, as well as gel consistency, peak viscosity, hot viscosity, and cool viscosity,while negatively correlated with globulin, gliadin, glutenin, and protein contents. The types of S/CRF and fertilization modes are important for improving rice yield and quality. Compared to CK, higher yield and similar quality of rice was achieved with UF as basal and CU as tillering fertilizer, and similar yield with improved appearance and eating and cooking quality of rice was achieved with either both UF and CU as basal fertilizer, or PCU as basal and CU as tillering fertilizer.

Keywords: type of slow or controlled release fertilizer, fertilization mode, yield, quality

1. lntroduction

Nitrogen (N) is an essential nutrient for all plants, and the application of N fertilizer plays an important role in rice(Oryza sativa L.) yield and quality (Ding et al. 2014; Jiang et al. 2016). To achieve high yield, quality, and N use efficiency in rice, numerous N application methods, such as precise quantitative cultivation techniques (Ling et al.2005), site-specific N management (Peng et al. 2010),and integrated soil-crop system management (Zhang et al.2011), have been developed through optimization of N input quantity and adjustments of application time and mode.However, in each of these methods, N fertilizer is typically divided into three or four broadcast applications (Peng et al. 2006). Multiple applications require additional time and labor, which will not meet the high efficiency demand of modern rice production.

Slow or controlled release N fertilizer (S/CRF) contain N in a form which delays availability for plant uptake post-application, thus eliminating the need for multiple applications (Trenkel 1997). Some research indicates that using S/CRFs can decrease NH4+concentrations in the surface water (Ji et al. 2006; Xu et al. 2012) and soil solution(Chen et al. 2011), which could lead to reduced loss through ammonia volatilization (Li et al. 2008; Wang et al. 2011; Lv et al. 2012; Zhang et al. 2012; Xu et al. 2013). Compared with common urea, N released from S/CRFs is synchronized to crop requirement patterns in the field (Dong and Wang 2006). The continual release of N from S/CRFs ensures adequate N supply for plant uptake throughout the growing season. Therefore, use of S/CRFs in rice production can lead to increased yield (Cao et al. 2005; Chen et al. 2012;Wang et al. 2015), increased N use efficiency (Chen et al.2005, 2010; Geng et al. 2015), improved root structure(Peng et al. 2013), increased activity of nitrate reductase and glutamine synthetase (Nie et al. 2003; Yang et al. 2012),and delayed leaf senescence in late growth stages (Tang et al. 2007). At the same time, use of S/CRFs have reduced greenhouse gas emissions including methane and nitrous oxide (Wang B et al. 2014) compared to use of common urea in rice paddies (Wang et al. 2016), thus having a lower impact on the environment (Ni et al. 2011).

According to the production process, S/CRF can be classified into two main groups. One is condensation products of urea and urea-aldehydes (slow release fertilizers) and the other is a coated or encapsulated fertilizers (controlled-release fertilizers) (Trenkel 1997).According to previous research, rice growth varied with use of different S/CRFs (Yang et al. 2013; Peng et al. 2014; Li et al. 2015). Compared to sulfur-coated urea, polymercoated urea releases N more slowly, which can dramatically increase above-ground biomass and grain yield with the continuous N supply, particularly at the late growth stage(Li et al. 2013; Wei et al. 2017). Furthermore, Xing et al.(2015) reported that use of a mixture of sulfur and polymer coated controlled release urea led to enhanced rice yield,larger leaf area index, and higher photosynthetic potential compared to use of these fertilizers individually.

Although the effects of S/CRF on rice yield, N absorption and N utilization have been thoroughly studied, little is known about the influence of fertilizer type and fertilization mode on rice quality. With improvements in living standards andstructural reform in the supply-side of agriculture, more high quality rice, especially in terms of eating quality, is needed in China (Peng et al. 2009; Calingacion et al.2014). To achieve these comprehensive goals (i.e., steady yield, high quality, reduced labor, environmentally friendly),improved fertilization methods will be essential. In thisstudy, experiments were designed to study the effects of S/CRF type and fertilization mode on rice yield and quality.These experiments included three types of S/CRF (polymercoated urea (PCU), sulfur-coated urea (SCU), and urea formaldehyde (UF)) and two fertilization modes (both S/CRF and common urea (CU) as basal fertilizer, S/CRF as basal and CU as tillering fertilizer). Results from this study may lead to recommendations for improving rice yield and quality through N fertilizer management.

2. Materials and methods

2.1. Experiment location and weather conditions

Field experiments were conducted during the rice growing season (May–October) in 2014 and 2015 at the agriculture experiment station of Yangzhou University in Heheng Village of Jiangsu Province, China (Shengao Town,Taizhou County). The field soil was a clay loam with 0.22%total N, 92.34 mg kg-1alkali hydrolysable N, 5.81 mg kg-1Olsen-P, and 163.02 mg kg-1exchangeable K. The daily mean temperature (Fig. 1-A) and sunshine hours (Fig. 1-B)during the growing season of rice measured at a weatherstation close to the experimental site are shown in Fig. 1.

2.2. Plant materials, growth conditions, and treatments

Japonica rice cultivar Nanjing 9108, which is the mostpopular rice with good eating quality in the middle and lower reaches of Yangtze River, was used as material. Seedlings were raised in plastic plates, sown on 31 May, both in 2014 and 2015 with a seeding rate of 120 g of dry seeds per plate.Seedlings were mechanically transplanted in hills on 20 June after wheat straw returning to the filed. Hill spacing was 11.7 cm×30 cm with four seedlings per hill. Three fertilizer types and two fertilization modes were tested in 25 m2(5 m×5 m) plots with three replications each treatmentmode combination. Each plot was separated by a soil ridge(35 cm wide and 20 cm high) and covered with a plastic film.

The three S/CRF types were PCU (21% N), SCU (37%N), and UF (16% N), and the two fertilization modes were 1) both S/CRF and CU applied as basal fertilizers, and 2)S/CRF applied as basal fertilizer and CU applied 7 days post-transplant at the tillering growth stage. Treatments were applied at a rate of 270 kg ha–1N in an S/CRF to CU ratio of 5:4. An additional control treatment (CK) was included with CU applied in four splits: 30% at basal, 30%at 7 days after transplanting, 20% as spikelet promotion fertilizer when the rice has four leaves that have not appeared, and 20% as spikelet development fertilizer when the rice has two leaves that have not appeared respectively.No S/CRF was used in control plots. Fertilizers of 150 kg ha–1P (as superphosphate) and 150 kg ha–1K (as KCl)were also applied and incorporated prior to transplanting.All fertilizer types, rates, and fertilization modes are listed in Table 1. The experimental field was flooded post-transplant and remained flooded until 7 days before maturity stage.Insect pests, pathogens, and weeds were controlled using common chemical treatments.

2.3. Sample and data collection

Rice yield was determined from a harvest area of 6.0 m2in each plot and adjusted to a standard moisture content of 14%.

Fifty hills of rice plants in each plot were harvested at maturity stage. Rice grains were collected, dried, and stored for more than 3 months according to NY/T83-1988 (1988).Grain samples of 120 g with three replications from each plot were collected for grain quality analysis according to GB/T 17891-1999 (1999). The length, width, and thickness of 10 rice grains including husk were measured with a vernier caliper. Samples were passed through a dehusker to get brown rice which was polished to get milled rice. Milled rice grains with grain length greater than or equal to 4/5 of its total length was manually separated to get head rice. The brown rice rate, milled rice rate, and head rice rate were expressed as the percentages of their weights to total rough rice weight(120 g). One hundred milled grains per plot were selected at random to check appearance quality. Grains containing a white belly, center, and back, or any combination of these,were considered chalky kernels. Milled rice was prepared to test amylose and starch content and gel consistency by grinding into flour with a stainless steel grinder and sifting with a 0.25-mm sieve.

Fig. 1 Daily mean temperature (A) and sunshine hours (B) during the growing season of rice in 2014 and 2015.

Table 1 Nitrogen fertilizer types, rates, and fertilization modes (kg ha–1)

Protein components were separated and analyzed following methods described by Ju et al. (2001). Albumins,globulins, glutelins, and prolamins were extracted from rice powder by treating with water, 5% NaCl, 0.02 mol L–1NaOH,and 70% alcohol, respectively. Albumin, globulin, and glutelin were precipitated from supernatants by adjusting the pH to their respective isoelectric points. Protein content was determined by the Kjeldahl Method, and crude protein content was calculated by multiplying the measured values by 5.95.

A total of 30 g samples of milled rice were cooked with a rice to water ratio of 1:1.3. Appearance, hardness, quality of balance, viscosity, and taste value of cooked rice were measured with the Satake Rice Taste Analyzer (STA1A,Satake Corporation, Japan).

Rice paste properties were determined using a Rapid Visco Analyser (RVA, Super3, Newport Scientific, Australia),following the procedure of the American Association of Cereal Chemists. A total of 3 g samples of flour were sifted with a 0.15-mm sieve and mixed with 25 g of deionized water in a RVA sample can. Peak viscosity, hot viscosity, cool viscosity in centipoise units (cP), and their derivative parameter breakdown (peak viscosity minus hot viscosity), setback (cool viscosity minus peak viscosity),and consistency (cool viscosity minus hot viscosity) were recorded with matching software of Thermal Cline for Windows (TCW).

2.4. Statistical analysis

Data were analyzed by analysis of variance (ANOVA) with SPSS 13.0 for Windows. Means were compared by the leastsignificant difference (LSD) test at the 0.05 probability level.

3. Results

3.1. Yield

Grain yields from S/CRF treatments showed a trend of UF＞PCU＞SCU, where UF was 5.00–6.96% higher than PCU and 12.34–13.77% higher than SCU (Fig. 2). Grain yields from fertilization mode of S/CRF as basal and CU as tillering fertilizer were 5.58–8.02% higher than those in the mode of both S/CRF and common urea (CU) as basal fertilizer.

Appropriate type of S/CRF and fertilization mode are important for high yield production of rice. Rice yields from treatments of B (PCU)+T (CU) (polymer-coated urea as basal and common urea as tillering fertilizer) and B (UF+CU) (urea formaldehyde as basal and common urea as tillering fertilizer) were similar to the CK, whereas treatment B (UF)+T (CU) (urea formaldehyde as basal and common urea as tillering fertilizer) was 5.2 and 5.9% higher than the CK in 2014 and 2015, respectively. Rice yield of other treatments including B (PCU+CU) (both polymer-coated urea and common urea as basal fertilizers), B (SCU+CU)(both sulfur-coated urea and common urea as basal fertilizers), and B (SCU)+T (CU) (sulfur-coated urea as basal and common urea as tillering fertilizer) were 6.62–12.14%lower than that of CK in 2014 and 2015.

3.2. Grain shape

Fig. 2 Effects of slow or controlled release fertilizer types and fertilization modes on rice. B (PCU+CU), both polymer-coated urea and common urea as basal fertilizers; B (SCU+CU), both sulfur-coated urea and common urea as basal fertilizers; B (UF+CU),both urea formaldehyde and common urea as basal fertilizers; B (PCU)+T (CU), polymer-coated urea as basal and common urea as tillering fertilizer; B (SCU)+T (CU), sulfur-coated urea as basal and common urea as tillering fertilizer; B (UF)+T (CU), urea formaldehyde as basal and common urea as tillering fertilizer; CK, common urea applied separately. Different letters indicatestatistical significance at the P=0.05 level. Bars mean standard deviation.

Among S/CRF types and fertilization modes, rice grain was the longest in the B (UF)+T (CU) treatment, and the ratio of grain length to width was the largest in the CK (Table 2).Grain width was influenced by fertilization mode. Within the same type of S/CRF, there were no differences in grain width or thickness between the two fertilization modes.

At last he felt convinced, in his own mind, that he really had one, and was so delighted that he positively17 died of joy at the thought of having at last caught an idea

Grain length and the grain length to width ratio had interactive effects between fertilization mode and fertilizer type. Meanwhile, interactive effects were also observed among year, fertilization mode, and fertilizer type on grain width and the grain length to width ratio.

3.3. Rice milling quality

Among S/CRF types and fertilization modes with the same level of total N, no differences were observed in the rate of brown rice, milled rice, or head rice (Table 3).

3.4. Rice appearance quality

Compared to the CK, appearance quality of milled rice was better from S/CRF treatments (Table 4). The average chalky kernel rate, chalky area, and overall chalkiness of milled rice from S/CRF treatments were 20.13, 22.97, and 37.77% lower than that of the CK, respectively, in combined results from 2014 and 2015.

Fertilizer type and application mode had effects on appearance quality of milled rice. Chalky kernel rate,chalky area, and overall chalkiness of different types of S/CRF showed the trend of UF＞PCU＞SCU. Within the same kind of S/CRF, chalky kernel rate, chalky area, and overall chalkiness of milled rice from the fertilization mode with basal applications of both S/CRF and CU were 11.26–22.61%, 9.17–25.55%, and 15.97–35.79%, lower than those from the fertilization mode with S/CRF as basal and CU as tillering fertilizer in combined results from 2014 and 2015. Meanwhile, interactive effects between fertilization mode and fertilizer type, as well as among year, fertilization mode, and fertilizer type on the appearance qualities of rice were also observed.

3.5. Rice nutritional quality

The average total protein and average globulin, gliadin,glutenin of milled rice from all S/CRF treatments were 8.50, 15.88, 12.15, and 9.75% lower than those of CK in combined results from 2014 and 2015 (Table 5). Fertilizer type effected the contents of gliadin, glutenin, and total protein in milled rice. Gliadin, glutenin, and total protein content of different S/CRF treatments showed the trend of UF＞PCU＞SCU. Within the same kind of S/CRF, gliadin,glutenin, and total protein contents of milled rice from the application mode with S/CRF as basal and CU as tillering fertilizer were 8.60–17.13%, 3.36–12.52%, and 3.67–8.10%higher than those from the application mode with both S/CRF and CU as basal fertilizer, respectively in combined results from 2014 and 2015. The interactive effects between fertilization mode and fertilizer type, as well as among year,fertilization mode and fertilizer type on the content of gliadin,glutenin, and protein were observed.

Table 2 Effects of slow or controlled release fertilizer types and fertilization modes on grain shape

Table 3 Effects of slow or controlled release fertilizer types and fertilization modes on rice milling quality

Table 4 Effects of slow or controlled release fertilizer types and fertilization modes on rice appearance quality

3.6. Milled rice starch content and gel consistency

The average amylose, amylopectin, and starch contents as well as the gel consistency of milled rice from all S/CRF treatments were 17.57, 1.14, 3.44, and 6.71% higher than those of CK in 2014 and 2015, respectively (Table 6). Among different S/CRF treatments, the amylose, amylopectin, andstarch contents of milled rice were the highest from the B (SCU+CU) treatment and the lowest in the B (UF)+T(CU) treatment. No differences in the gel consistency were observed among S/CRF treatments. Within the same kind of S/CRF, the amylose, amylopectin, and starch contents in the modes of both S/CRF and CU as basal fertilizer were 7.63–18.80%, 1.10–1.47%, and 2.14–3.97% higher than those in the mode of S/CRF as basal and CU as tillering

fertilizer in 2014 and 2015. Interactions between fertilization mode and fertilizer type, as well as among year, fertilization mode and fertilizer type, impacted amylose and starch contents of milled rice.

Table 5 Effects of slow or controlled release fertilizer types and fertilization modes on rice nutritional quality

Table 6 Effects of slow or controlled release fertilizer types and fertilization modes on starch content and gel consistency of milled rice

3.7. Rice eating quality

3.8. Rice starch viscosity characteristics

The peak viscosity, hot viscosity, cool viscosity, breakdown,and consistency of milled rice from S/CRF treatments were higher than those in CK with the exception of the B (UF)+T (CU) treatment (Table 8). Within the same kind of S/CRF,the peak viscosity and hot viscosity of milled rice from the mode of both S/CRF and CU applied as basal fertilizers were 2.45–9.40% and 2.32–30.39% higher, respectively,than those of the mode with S/CRF as basal and CU as tillering fertilizer in 2014 and 2015.

3.9. Correlations between protein components,starch composition, and eating quality

Globulin, gliadin, glutenin, and protein contents were negatively correlated to starch composition, appearance,quality of balance, viscosity, and taste value (Table 9). At the same time, high globulin, gliadin, glutenin, and protein content correlated to increased hardness and decreased appearance, quality of balance, viscosity, and taste value of cooked rice. Amylose, amylopectin, and starch content were all positively correlated with appearance, quality of balance,viscosity, and taste value, while negatively correlated with hardness of cooked rice.

Table 7 Effects of slow or controlled release fertilizer types and fertilization modes on rice eating quality

Table 8 Effects of slow or controlled release fertilizer types and fertilization modes on rice starch viscosity characteristics

3.10. Correlation between starch viscosity and eating quality

Gel consistency, peak viscosity, hot viscosity, and cool viscosity were all positively correlated with appearance,viscosity, and taste value, while the peak viscosity and breakdown were negatively correlated with hardness in cooked rice (Table 10).

4. Discussion

4.1. Effects of slow or controlled release fertilizer types and fertilization modes on yield of rice

The results of this study suggest that selection of S/CRF type and fertilization mode can positively impact rice yield and quality while also saving associated labor costs. However,the effects of S/CRF type and fertilization mode on rice yield and quality differed by the mechanism of N release (i.e., slow or controlled). The polymer or sulfur coated urea fertilizers used in this study have a protective (water-insoluble)coating around the urea nucleus. Nitrogen release from coated controlled release fertilizers is mainly controlled by the thickness of the coating material, and is also affected by temperature and humidity (Du et al. 2003; Peng et al.2014; Wang S P et al. 2014). UF fertilizer is formed by the reaction of formaldehyde with excess urea under controlled conditions, resulting in a mixture of methylene urea with different long-chain polymers. Nitrogen release from UF is influenced by the length of polymer chains as well as soil organisms and their activity.

Compared to polymer or sulfur coated urea, N release from UF was more stable and effective at high temperature and humidity in the south area of China. The treatment of UF achieved a high yield of rice with large number of spikelet per area, large leaf area, and more biomass accumulation particularly at late growth stage. At the same time, the rapid initial N release of sulphur coated urea (Peng et al. 2013; Xu et al. 2016) may have led to lower rice yield compared to that of polymer coated urea. Due to a delay in the availability of N, S/CRFs are commonly used in combination with common urea to meet the N requirement at early stage of rice (Ke et al. 2017; Zhang et al. 2017). Treatments with basal S/CRF and CU at tillering stage may have had an increased number of productive tillers and spikelets compared to treatments with basal SRF and CU (Wei et al. 2017), which may result in a higher yield of rice, as found in this study.

4.2. Effects of slow or controlled release fertilizer types and fertilization modes on quality of rice

The effects of N fertilizer on the indexes of rice quality were different. Previous research reveals that the appropriate amount of N fertilizer can decrease the chalky kernel rate(Zhou et al. 2015) and overall chalkiness (Alcantara et al.1996), while overuse of N can increase the chalky kernel rate and undesirable grain appearance (Ma et al. 2010; Xi et al.2016). Among different S/CRF types and fertilization modes,chalky kernel rate, chalky area, and overall chalkiness showed the tend of B (UF)+T (CU)＞B (PCU)+T (CU)＞B(SCU)+T (CU)＞B (UF+CU)＞B (PCU+CU)＞B (SCU+CU)(Table 4). Possible reasons for this finding include: 1)delayed release and function of N in the type of S/CRF(such as UF) (Alcantara et al. 1996; Huang et al. 2016),or 2) the fertilization mode of S/CRF as basal fertilizer and CU as tillering fertilizer, led to an increase in the number of spikelets per area while decreasing the rate of seed set in rice (Wei et al. 2017). At the same time, the appearance quality tends to deteriorate due to poor rice grain filling (Ma et al. 2010). Similar results were also found by Zhao (2014)where an increased rate of panicle fertilizer resulted in higher frequency of rice grains with a white belly.

Eating and cooking quality of rice can be determined by several factors including the content of amylose and protein,gel consistency, and starch viscosity (Ramesh et al. 2000;Aluko et al. 2004; Tian et al. 2005; Bao et al. 2006a, b).These indexes were all influenced by the application of N in this study. Taste value of different treatments showed the trend of B (SCU+CU)＞B (PCU+CU)＞B (UF+CU)＞B (SCU)+T (CU)＞B (PCU)+T (CU)＞B (UF)+T (CU). Amylose,amylopectin, and starch contents were all negatively correlated with the content of albumin, globulin, gliadin,glutenin, and protein (Table 9). With different treatments of S/CRF types and fertilization modes in a single cultivar of rice, the relatively improved metabolism of N tends to restrain the metabolism of carbon during grain filling(Qiao 2011). Therefore, the certain type of S/CRF or fertilization mode with sustained release of N, particularly at the late growth stage of rice, tended to increase protein components while decreasing amylose, amylopectin, andstarch contents. Starch wrapped with more protein in rice tended to result in the firmer texture of cooked rice (Martin and Fitzgerald 2002) with shorter gel consistency and lower viscosity. That may be why the eating and cooking quality from the B (UF)+T (CU) treatment was lower than that from the B (SCU+CU) treatment.

5. Conclusion

Rice grain yield, chalky kernel rate, chalky area, overall chalkiness, and the contents of gliadin, glutenin, and protein all show the trend of UF＞PCU＞SCU within the same fertilization mode, and the trend of S/CRF as basal and CU as tillering fertilizer＞both S/CRF and CU as basal fertilizer within the same kind of S/CRF. While the content of amylose, amylopectin, and starch, the taste value, and the peak and hot viscosity show the trend of SCU＞PCU＞UF within the same fertilization mode, and the trend of both S/CRF and CU as basal fertilizer＞S/CRF as basal and CU as tillering fertilizer within the same kind of S/CRF.Compared to CK, higher rice yield with similar quality can be achieved with the B (UF)+T (CU), and similar yield with improved appearance, eating and cooking quality of rice can be achieved using either B (UF+CU) or B (PCU)+T (CU).

Acknowledgements

We are grateful for grants from the National Key R&D Program of China (2016YFD0300503), the Key Research Program of Jiangsu Province, China (BE2016344), the National Rice Industry Technology System, China (CARS-01-27), the National Nature Science Foundation of China(31701350), the Program for Scientific Elitists of Yangzhou University, China and a Project Funded by the Priority Academic Program Development of Jiangsu Higher Education Institutions, China.