Soil nutrition absorbed by crops can be divided into mobile and immobile [1] . Nitrogen (N) in form of nitrate and water are highly mobile and required in largest amounts by crops. Phosphorus (P) is the most immobile, and potassium (K) is also relatively immobile, both of which are macronutrients required by crops [2] . The contents of N, P and K in agricultural soil are affected by plant growth and yield [3] . Therefore, crop yield is limited by two important mobile resources, including nitrate and water, as well as two immobile resources, P and K [4] . In recent years, there was about 60% of soil nutrition deficiency as a result of long-term agricultural production of existing cultivated land in China, could not meet with the needs of crop yield improvement [5] .

Fertilizer plays an important role in crop yield improvement, which increased crops grain yield by 55% - 57%, and contributed to 30% - 31% of the total grain production [6] [7] . The nitrate N is easily lost through leaching and denitrification in field soil, whereas the ammonium N is usually lost through volatilization [8] [9] . China is a big country in consumption of N fertilizer [10] . The effects of N fertilizer applications on soil organics matter status and soil physical properties are importance to agricultural sustainability and to increase crop yield [11] . Modern agriculture cultivation, on the contrary, concentrates on supper high grain yield and maximum output, compromising input-use efficiency, therefore may not be sustainable in the long run.

Maize (Zea mays L.) is one of the most important food crops in the world. N is one of essential nutrient elements for maize growth and development, which use 1 kg of N to produce 49 kg of grain [12] . Although more N fertilizer has been applied, N use efficiency has turned lower. The investigation indicated that utilization rate of N fertilizer was 20% to 50% in china [13] [14] . In addition, more and more application of N fertilizer caused pollution of groundwater and other problems [15] [16] . The extreme of excess application and N pollution were found in intensive agricultural systems of Western Europe, the United States, and, more recently, China [17] . Therefore, reasonable field managements and appropriate application of N fertilizer are necessary for the supper high yield of maize.

Many approaches have been practiced for improving N utilization efficiency in crops, for example, optimal time, rate, and methods of application for matching N supply with crop demand and the use of specially formulated forms of fertilizer. The results showed that N application by stages can significantly increase maize grain yield compared to disposable application as sowing manure [18] . Zhang et al. (2014) reported that the regulating N application (240 kg/ha, divide into 3 equal amounts, each about 80 kg, used as base fertilizer, tillering fertilizer, and booting fertilizer) could increase rice yield while substantially reduced N leaching losses and improved N use efficiency in the upper reaches of the Yellow River, China [19] . At present research on application of N fertilizer roughly includes onetime application technique, basal application and side dressing, the amount of N fertilizer application, slow-released fertilizers and so on. However, there are few researches on regulating N application for maize.

In present study, nine N treatments were carried out to evaluate the effect of different nitrogen applications on soil physical, chemical properties and grain yield production. The objectives are to evaluate the effects of different N fertilizer application on soil physical and chemical characters and maize yield, and to identify the approach for optimal N fertilizer application in maize management program.

The experiment was carried out on an experimental farm located in Daguben town (N 42˚28′, E 122˚22′), Fuxin city, Liaoning province, China. The hybrid maize variety Zhengdan 958, widely cultivated in northeastern of china, was used in this study. Nine N fertilizer treatments were arranged in a randomized complete-block design with three replicates, totally including 27 plots. Planting were cultured on 30^th April, 2011 and on 6^th May, 2012. And each plot was 5 m wide by 8 m long with 10 rows, row spacing 50 cm .

The nine N treatments received T₁ (N 0 kg/ha), T₂ (CK, compound fertilizer 108.75 kg/ha, N 29%, P 10% and K 11%), T₃ (N 138.0 kg/ha, 30% at sowing and 70% as side-dressing at jointing stage), T₄ (N 241.5 kg/ha, 30% at sowing and 70% as side-dressing at jointing stage), T₅ (N 345.0 kg/ha, 30% at sowing and 70% as side- dressing at jointing stage), T₆ (N 241.5 kg/ha, 20% at sowing, 60% as side-dressing at jointing stage and 20% at big flare period), T₇ (N 241.5 kg/ha, 30% at 7 cm soil layer and 70% at 15 cm soil layer), T₈ (N 205.2 kg/ha, Jin zhengda slow-released urea of N 35% at 15 cm soil layer), T₉ (N 241.5 kg/ha, Jin zhengda slow-released urea of N 35% at 15 cm soil layer), respectively. In seven treatments from T₃ to T₉, phosphorus (P₂O₅ 103.5 kg/ha) and potassium (K₂O 144 kg/ha) fertilizers were ploughed into the soil tillage layer in one time as a basal fertilizer.

Soil samples were collected from a depth of 0 - 20 cm on the ridge after harvest in 2011 and 2012. Soil bulk density, soil porosity and field moisture were measured by Wilcox method. Soil organic matter content was measured by potassium dichromate method. The total N content of soil and plants were calculated by using the Kjeldahl N method, total P content using Mo-Sb colorimetric method, and total K content using flame photome- try described by Zhang et al., 2014 [19] . The pH was measured by composite electrode method.

Plant samples were taken at jointing stage and big flare period from the two center rows of each plot. Grain yield were determined by harvesting the two center rows from each plot. One way analysis of variance (ANOVA) at α = 0.05 probability was conducted to test the significance in different treatments.

Soil bulk density is the ratio of the mass of dry solids to the bulk volume of the soil. The bulk volume includes the volume of the solids and of the pore space. The bulk density reflect the compaction soil, and influence the transform and utilization rate of nutrient in soil directly [20]. In the present study, the soil bulk density of T₁ and CK were lower than other treatments in 2011 and 2012, whereas the soil bulk density of T₅ in 2011 and T₃ in 2012 were higher than other treatments. Field capacity is the amount soil moisture held in soil after

excessive water has drained away and the upper limit of water stage [21] . Soil porosity and assignment can be affected by the bulk volume [22] . The soil porosity and field capacity of T₂ in 2011 and T₁ in 2012 were higher than other treatments. On the contrary, those of T₅ in 2011 and T₃ in 2012 were lower than other treatments.

Zhong and Shangguan (2014) reported that N-applied treatments increased water consumption in different layer of soil and evapotranspiration, which were significantly higher in N-applied than in non-N treatments [23] . The average soil pH has decreased 0.5 units due to the excess utilization of N fertilizer in the past two decades in China [24] . Li et al. (2013) reported that the soil pH declined from 8.76 to 8.56 from 1992 to 2008 during long-term field trials in North region, China [25] . The pH values of T3 to T7 were lower than other treatment in this study. These results indicated that N application by stages could increase the soil bulk density and decreased the soil pH, while non-N fertilizer or slow-released urea application should lead to soil harden.

N uptake and efficiency utilization by maize is very important to N economy and yield improvement in agricultural production systems [26] . The dry matter production and nutrient accumulation were usually positively correlated with crop grain yield [27] . Meng et al. (2013) reported that the wheat yield increased from 7 - 9 Mg∙ha⁻² to > 9 Mg∙ha⁻² was mainly attributed to increased dry matter and N accumulation from stem elongation to anthesis period in eleven filed experiments [28] . Ma and Dwyer (1998) indicated that prolonged maintenance of green leaf area for photosynthate and the ability to take up available soil N during grain filling were characteristics of hybrids maize with greater NUE [29] . Zhang et al. (2013) reported that the contribution of remobilized N from maize organs to grain showed a trend of blade > stem and sheath > cob > bract according to the maximum value of accumulated N in organs [30] . In present study, N contents of shoot were not significant different among nine treatments at joint stage. However, N contents of leaf in T₃ to T₇ were significant higher than T₁, T₂, T₇ and T₈ at big flare period. Specially, N content of leaf in T₄ and T₅ was the higher than other treatments. Al- though the N, P, K contents of leaf and stem in T₅ treatment were higher than other treatments, yield of T₅ was not the highest. These results indicated that N application should increase the N content of leaf and stem, but only reasonable level should increase N utilization efficiency and significant improve maize yield. These par- tially attribute to excessive growth caused by overuse fertilizer and vegetative growth consumed more nutrition during growth period in crops.

Soil organic matter content is a major source of system stability in agro ecosystems. Soil total N and alk-hydr. N contents are important fertility indexes of soil. Zhou et al. (2013) reported that the soil organic carbon and total nitrogen concentrations had a significant effect on crop yield in the semi-arid Loess Plateau by long-term experimentation [31] . Gong et al., (2013) also indicated that the contribution of soil productivity was significantly correlated with soil organic carbon, total nitrogen, available nitrogen, available phosphorus and available potassium in wheat with long-term soil fertility experiments [32] . The same results were reviewed from the 1970s to the 2000s in the Loess Plateau in China by Wang et al. (2014) [33] . Results showed that changed trends of soil organic matter content under different treatments in two years were similar. Both soil organic matter and to- tal N content of T3 were the highest in both 2011 and 2012. However, the yield of T₄ was higher than other treatments in both two years. Bassoa et al. (2010) indicated that yield response was stronger for 120 kg N/ha than 60 kg N/ha and 90 kg N/ha with the long-term wheat experiment response to N under rain-fed Mediterranean environments [34] . So these results indicated that separated N application could significantly increase grain yield compared to only one application at sowing, but N application times should also be controlled. Not only the amount of N fertilizer application should be taken into account, but also N application periods and times should be controlled. In our study, control release urea didn’t significantly increase maize production and didn’t take a significant advantage in improvement of N use efficiency. But the results of control release urea application in 2012 were better than that in 2011. The advantage of control release urea should be obvious in some years.

In the present study, nine treatments of N fertilizer application were carried out to evaluate the variances of soil physical and chemical, the contents of N, P and K in plant and maize grain yield. Results indicated that the soil bulk densities were increased, whereas the soil porosity, field capacity and pH values were decreased with more N application. Reasonable N fertilizer amount (241.5 kg/ha) and application at two stages (30% at sowing and 70% at jointing stage) could significant increase N utilization efficiency and improve maize yield.

The work was financially supported by Key Projects in the National Science & Technology Pillar Program during the Twelfth Five-Year Plan Period (2011BAD16B12, 2012BAD04B03, 2013BAD07B03), and by the Tianzhu Mountian Scholars Support Plan of Shenyang Agricultural University, and by Program for Liaoning Excellent Talents in University.