Preparation of Nanoparticles Preparation of iron nanoparticles were done in the laboratory of Department of Agronomy and Agricultural Extension, University of Rajshahi, Bangladesh. In the present study, the aquostic method was used to prepare the metallic iron nanoparticles. A mixture of reagent (salt) and polymer surfactant in ethylene glycol (EG) and/ H2O was heated in an oil bath for several minutes and desired shapes and sizes of nanostructures in high yields were prepared by changing various experimental parameters such as concentrations of reagent surfactants (e.g. PVP), reluctant and solvents (EG or other polyols), gas bubbling, temperatures and heating rate. The synthesized nanoparticles were purified by the precipitation method. Crystal structures and growth mechanisms of nanoparticles were characterized by Atomic Force Microscopy (AFM) (Park system.XE-70, South Korea). Produce solutions were centrifuged at 6,000 rpm three times for 30 min to ensure complete collection of products catch time. The precipitates were collected and then re-dispersed in ethanol. Samples for AFM measurements were prepared by dropping a droplet of the colloidal solution on the glass slides.
All feed ingredients were purchased from local market and in laboratory they were grinded to acquire fine powder. The powdered and sieved feed ingredients were weighed out and mixed thoroughly in 6 different ratios for preparing six different diets 1 control and 5 different diets containing Fe-NPs at various doses such as 0 (control), 10 (Fe-NPs10), 20 (Fe-NPs20), 30 (Fe-NPs30), 40 (Fe-NPs40) and 50 (Fe-NPs50) mg/kg dry feed weight. Then distilled water was added and blending well (5 min) until the mixture achieves a dough consistency. The dough was pelletized in a manual pelletizer fixed with 3 mm diameter and the pellets were collected in aluminum trays. A thermostatic hot air oven (Microsil INDIA, Universal Lab Product Co., Chennai, India) was used to dry the diets until the moisture content was reduced below 10%. After drying diets were kept at 20ºC until used.
Collection and maintenance of fish Juveniles of C. batrachus having an average weight of 5.23±0.07 g were purchased from Fish Seed Hatchery, Rajshahi and transported live in aerated plastic bags to the laboratory of Department of Fisheries, University of Rajshahi. Fishes were kept in a cemented tank having flow through system and were acclimatized for two weeks. During the acclimatization water quality parameters maintained in the optimum range temperature, 27-30ºC, with a photoperiod of 12-h light and 12-h darkness.
Experimental Design After an acclimatization period, healthy and uniform sized fishes were selected, individually weighed by using electronic top-loading balance and evenly distributed in eighteen fiber glass aquaria at 10 fish per aquarium with similar initial weight. The experiment was conducted as a Completely Randomized Design (CRD) with six treatments as control, FeNPs10, Fe-NPs20, Fe-NPs30, Fe-NPs40 and Fe-NPs50 each with three replications. Fishes were fed daily (twice a day) with a feeding rate of 3% body weight. After feeding period, the diet remaining in each tank was collected by siphoning before the second day’s feeding. A routine work of exchanging 50% water from each aquarium was done daily.
Water quality analysis Water temperature was measured using a Celsius thermometer. Water pH was measured using an electronic pH meter (Jenwary, 3020). Dissolved oxygen (DO) and ammonia were measured by using a portable aquaculture kit (Model FF2, HACH, USA).
Muscle Composition At the end of the study period, two fishes were randomly collected from each aquarium and were used for determination of proximate composition in laboratory of Department of Applied Chemistry, University of Rajshahi, Bangladesh. Fish tissue samples were dried and one gram of dried tissue samples was digested separately with 10 ml of HNO3 in a microwave device (CEM MDS 2100). Following digestion, the samples was diluted with distilled water to make up 20 ml and filtered. Crude protein was estimated by micro- kjeldahl method; crude lipid by petroleum ether extraction using the soxhlet method [12]; moisture and ash content were analyzed following the method of AOAC [13] . Muscle and liver iron content were determined by Flame Atomic Absorption Spectrometer (Shimadzu, AA-6800) in central lab of University of Rajshahi, Rajshahi, Bangladesh followed by digestion in Bangladesh Council of Scientific and Industrial Research (BCSIR), Rajshahi, Bangladesh.
Statistical Analysis: The data were analyzed by one-way analysis of variance (ANOVA) using SPSS (20.0), followed by Duncan’s multiple range test to compare the differences among treatments where significant differences (P< 0.05) were observed. Data were expressed as mean ± SD.