A research in order to determine the initial feasibility to generate biohydrogen from urban organics wastes and then established some conditions to operate a batch bioreactor was developed in Colombia. This section presents the results of this research and analysis the potential use of urban wastes as sources to generate hydrogen.

1.2. Localization

The research was performance between the years 2009 and 2012, at the Laboratory of Agri­cultural Mechanization of the National University of Colombia in Medellm, localized in 6°13’55"N and 75°34’05"W, with average annual temperature of 24°C, relative humidity of 88% and average annual precipitation of 1571mm.

1.3. Methods

Two stages were established to develop the research, the first had five phases.

1.3.1. First stage

Phase 1. Identification of organic wastes generated at the Central Wholesaler of Antioquia

The Central Wholesaler of Antioquia is the main company dedicated to trade food in the city of Medellin (fruits, vegetable and some grains). At the first phase historical information related to organic wastes production during two year was supplied by Central Wholesaler of Antio- quia and was made a photographic register of solids wastes generated. The photographs were taken twice per day at the morning and afternoon.

Phase 2. Selection of wastes with greater production

According to the information collected and the photographic register from the first phase, the wastes with greater production were selected to be introduced into a batch bioreactor.

Phase 3. Elemental Composition and chemical composition analysis

The quantity of volatile solids, total solids and elemental composition on both wet and dry basis (coal, nitrogen and hydrogen) were obtained for each wastes. Were taken samples of 5 grams and the analysis method applied was the Wendee method (the analysis was made at the chemical analysis laboratory of National University in Medellin). With that information was calculated the quantity of wastes to use. Six samples of 3 grams in each wastes were taken in order to obtain the elemental analysis, in this case the method applied was burn of sample and the equipment employed was an elemental analyzer CE — 440 (Figure 2a). The samples were triturated with a precision crusher — IKA WERNE with sieve of 0,5 mm (Figure 2b) and then were dried in a lyophilizer LABCONCO Freezone 12L (Figure 2c). In order to determine the quantity of wastes and water to be employed, 6 grams of volatile solids per liter-day were used as organics load [19], additionally was employed a concentration on volatile solids of 5% [20].

A batch bioreactor of 2000 liters was installed, the wastes were triturated to facilitate its access into bioreactor and its process by the bacteria. The quantity of gas generated was registered with a gas flow meter Metrex G 2,5 with accurate of 0,040 m3/h; maximum pressure of 40 kPa, additionally was employed a gel of silica to remove the wet of gas. The load of bioreactor was made during four days, each day was used the same quantity until to complete the total load.

Phase 5. Principal variables to register

The relativity humidity and environment temperature were registered daily, was used a thermohygrometer with rank in temperature until 120°C and 100% in relativity humidity (Figure 4). The pH into the bioreactor was registered daily too, in this case was employed a digital pH-meter Hanna Instruments, with accurate of ± 0,2 (reference temperature of 20°C).

Figure 4. Thermohygrometer and pH-meter

The organics load was determined at the beginning and end of bioprocess; in this case the total suspended solids (TSS), total solids (TS), volatile fatty acids (VFAs), chemical oxygen demand (COD) and biochemical oxygen demand (BOD) were determined. The analytics method employed were Standard Method by water and residual water of the APHA-AWWA-WPCF, edition 19 of 1995.

The production of gas was registered daily, samples were collected in Tedlar bags (with capacity of 1 liter, Figure 5) and then were analyzed in a chromatographic gas (Perkin Elmer) to establish its composition (percentage of CO2 O2 H2 CH4 and N2). During the tests, the wastes were subjected to an acid pretreatment to eliminate the methanogenic bacteria, after several days, agricultural lime was added to increase the pH until to obtain a value most adequate to the acidogenic bacteria.