Effect of blend ratio

A better understanding of the NO emissions can be observed when the fuel nitrogen content and the NO emitted are compared as shown in Figure 3.37. It can be observed that as the blend ratio is increased the fuel nitrogen content increases. Thus one would expect the formation of fuel NO to increase as the blend ratio is increased. On the contrary, the NO generated decreases. In the situation where the NO formation is only through fuel nitrogen, it is remarkable to observe that increasing FB content in cofired fuel does not necessarily lead to higher NO emissions. The trend is observed at different excess air ratios. Thus, the use of FB not only serves as a means of waste disposal, but also reduces the emissions. The reduction of NOX with coal:FB blends is due to better grindability of FB compared to fibrous DB.

3.8 REBURN

Previous sections dealt with co-firing of FB and DB with coal in conventional boiler burners. In this approach, the high temperatures produced by the coal allow for the successful combustion of the FB. A 30 kW Boiler Burner Facility was built at Texas A&M and co-firing was performed which revealed better combustion of coal with FB and similar or less NOX when cofired with coal even though FB has 2 and 4 times the N content of coal on a mass and heat basis. Small — scale co-firing tests were followed by pilot plant tests at the 500 kW facility of National Energy Technology Laboratory of DOE-Pittsburgh with similar results (Annamalai et al., 2003b, 2003c). Further, the VM of CB on DAF basis was almost twice that of coal. The large amount of volatile content of the biomass in the blend consumes oxygen rapidly in the near-burner region, thereby creating more localized fuel-rich zones, and hence less NO is formed. In addition, it is believed that most of the N in FB exists as NH3, the volatile matter of FB is twice that of coal and hence NOX emission did not increase. If so, the CB can serve as effective reburn fuel forNOx reduction. A premixed propane and trace amounts of NH3 were burnt to simulate coal combustion gases and use NH3 to produce NO in the main burner, and then test coal and feedlot biomass as reburn fuel.

The experiments were conducted in the Texas A&M laboratory scale boiler burner that was modified for reburn experiments. The boiler is a 30 kW (100,000 Btu/h) downward-fired furnace

made up of a steel shell encasing ceramic insulation. A schematic of the entire setup is shown in Figure 3.38. A premixed propane burner is mounted at the top of the furnace to produce hot furnace gases to simulate the products of coal combustion. Ammonia is injected into the premixed propane fuel stream and burnt in the primary zone. The primary or main burner zone equivalence ratio (фм) is typically <1 indicating excess O2. The reburn fuel is fed from a dry solids feeder, through a venturi inductor value, and injected through the reburn ports. The reburn injection ports are located below the tip of the premixed propane flame, after all of the NO has been formed in the primary zone. The reburn zone below the main burner receives excess O2 left from the main burner and the O2 from the reburn injector operated at equivalence ratio ф^. Thus the reburn zone equivalence ratio (фкв, ъ) is adjusted both by фм, фкв and fraction of thermal output (x) released by reburn as shown below.

Consider a reburn facility operated with total thermal rating of P [kW]. The main fuel M is fired at an equivalence ratio фм (< 1) such that it has a rated thermal output of PM = (1 — x)P. The reburn fuel R is fired at fR (> 1) such that its rated power output is PR = x P. The fR is adjusted so that the reburn zone equivalence ratio is ф^^ Then a relation between фкв can be obtained in terms of фм, ф^^ HHVO2,M and HHVO2,RB of main fuel and return fuel, and x, fraction of heat by reburn fuel:

image147

Figure 3.39. Required reburn injector equivalence ratio for desired reburn zone equivalence ratio for various reburn heat input %.

image148

Figure 3.40. Effect of reburn zone equivalence ratio on NOX reduction with various reburn fuels (reburn heat input = 30% of total thermal rating; main burner NO: 600 ppm) (adopted from Annamalai and Thien, 2001).

The 0KB was adjusted to obtain the desired equivalence ratio 0kbZ from 1 to 1.1 (Fig. 3.39). An Enerac 3000E gas analyzer is then used to measure the concentration of oxygen and NO in the final sampling port. After passing by the gas sampling port, the furnace gases are cooled by a water spray and exhausted out of the building. There is no burnout zone in the current boiler burner configuration. Figure 3.40 shows the results; it is seen that NOX reduction is highest for FB due to (i) increased VM of FM, which reduces local O2, (ii) release of N probably in the form of NH3.

Currently, experiments are in progress on coals, and CB to determine the percentage nitrogen distribution between HCN and NH3. However, we have recently analyzed data presented elsewhere (Di Nola, 2009), which demonstrated that adding animal waste to coal increased the ratio of NH3 relative to HCN. It is noted that the emissions of HCN and NH3 are not expected under combustion conditions.

The fuel N evolved as NH3 undergoes oxidation reaction with O2 and reduction reactions with HCN and NH3. The overall competing reactions for “thermal” (i. e. temperature dependent) De-NOx process is as follows:

2NO + 2NH3 + (1 /2)O2 ^ 2N2 + 3H2O, destruction of NO, 871-1204°C (1600-2200°F)

(3.40)

2NH3 + 2.5O2 ^ 2NO + 3H2O, oxidation of NH3, T > 1204°C (2200°F) (3.41)

The upper end of the temperature window is caused by the rapid growth of chain carriers which enhances reactions involving the oxidation of NH2 eventually producing NO instead of reducing it (Lyon et al, 1986). Sometimes the upper end could be as high as 1204°C (2200°F) (EPA). Exxon had empirically determined that NOx reduction is effective at T < 955°C (1750°F). Typically reactions are faster in the presence of O2 but not in excess amounts; these reactions suggest that the stoichiometric ratio of mole of NH3 to mole of NO is about 1; the actual amount of NH3 needed for the reaction is much greater than the theoretical amount because NH3 reacts with several other gases in the flue gases, not just NOx. The literature suggests that one needs about

0. 5-3 moles of NH3 per mole of NOx. The Selective Noncatalytic Reduction process (SNCR) temperature window is about 900°C to 1100°C. To facilitate quick and inexpensive predictions with the thermal De-NOx method, two competitive reaction formulations have been used for modeling purposes. One may use an empirically based model (Lyon, 1987), which includes the following forward direction only competitive reactions (Thien et al., 2012):

Reaction A: 4NH3 + 4NO + O2 ^ 4N2 + 6H2O (fast)

Reaction B: 4NH3 + 5O2 ^ 4NO + 6H2O (slow)

NH3 oxidation, >1480K

d(NO)/dt, NO production/reduction rate per unit volume, {kmol/(m3 s)} = kB (NH3) — kA (NH3) (NO)

d(NH3)/dt, NH3 consumption rate per unit volume, {kmol/(m3 s)} = — kB (NH3) — kA (NH3) (NO) where (NH3), (NO) concentrations in kmol/m3 and the specific reaction rate constants k’s are defined as

3 17 (— 29400

ka, {m3/(kmol s)} = 2.45 x 1017 exp(———- —— j, T inK

kB, {1/s} = 2.21 x 1014expl———- ——j, T inK

Since O2 is in excess typically and others are in trace amounts, reaction A depicts the second — order reduction of NO to N2 and the reaction B represents the first order oxidation of NH3 (Duo et al., 1992).

It can be seen from the values of the two activation energies and the overall rate constants how the simple model was able to predict the temperature window for NO reduction (1145 to 1480 K). Figure 3.41 shows typical model results for an initial NO of 600 ppm and assumed NH3/NO = 2 at 1100 K. The model results showed that when NH3/NO was set to 1, the rate of reduction slowed down and when NH3/NO = 0.5, the NOx reduction was only 50%. While pure FB produces a

Подпись: Effect of residence time on NOX reduction by NH3 at 1100 K.

Figure 3.41.

high amount of NH3, coal:FB blends result in lesser NH3 concentration; similarly lesser reburn heat input will result in lesser NOX (Oh, 2008).

It is apparent that coal and FB can both be successfully used a rebum fuel in order to reduce NO in a boiler burner. Feedlot Biomass is almost two times more effective as a reburn fuel than coal. The NO reduction is more effective at higher reburn equivalence ratios for coal; however, the NOX reduction is almost independent of equivalence ratio for feedlot biomass. The behavior of coal-biomass blends falls in between the behavior of coal and biomass. The greater effectiveness of feedlot biomass may be due to the release of fuel nitrogen in the form of NH3, and its high volatile content on a dry ash-free basis (Annamalai and Sweeten, 2005).