Skip to main content
SLU publication database (SLUpub)
Report, 2011

Glycerin från omförestring av vegetabiliska oljor som tillsatsmedel : praktiska försök med pelletering och eldning av några biobränslen

Bernesson, Sven; Örberg, Håkan; Samuelsson, Robert; Thyrel, Mikael; Hedman, Björn; Kalén, Gunnar

Abstract

During transesterification of vegetable oils in particular, but also animal fats and used frying oils, glycerine is obtained as a by-product. It is important for transesterification economics that the glycerine by-product can be sold for the best possible price. In Sweden today, glycerine is often sold to biogas production for a few Swedish crowns per kilo. If other markets where glycerine replaces a more expensive product can be found, the willingness to pay for glycerine will increase. This project examined whether glycerine can be used as an additive in pelleting some biofuel types, and the value it would have in this use. The project investigated admixtures of appropriate amounts of glycerine of varying quality during pelleting of some types of fuel (straw, reed canary grass and pine wood). To determine how these affect pellet quality, the function of the pellet press, pellet storage properties, the risk of harmful emission levels of carbon monoxide (CO), nitrogen oxides (NOx), sulphur oxides (SO2) and aldehydes during incineration of the pellets, the risk of ash sintering and the risk of corrosive deposits in flue gas channels were examined. In addition, the commercial value of the glycerine when added to pellets was estimated. Pelleting trials were conducted using 1% and 5% admixtures of four types of glycerine in three types of fuel. Two of the glycerine products were alkaline, originating from alkaline transesterification of rapeseed oil, and two were acidic, originating from acid esterification/transesterification followed by alkaline transesterification, of rapeseed oil in one case and used frying oil in the other. The fuel types were pine wood, reed canary grass and straw. After pelleting, pellet durability, the amount of fine fraction and bulk density were measured. Chemical analyses were performed of glycerine types, fuel types and glycerine/fuel mixtures to evaluate ash melting behaviour and the risk of corrosion associated with alkali chlorides. Pellets from the trials were stored for 6 months, during which time moisture uptake and the incidence of mould were studied. After the storage experiments, pellet durability was measured again. Wherever possible, the experimental design was evaluated statistically using multivariate data analysis. About half the samples were incinerated and emissions of CO, NOx, SO2, particulate matter, acrolein and formaldehyde were measured. In addition, the average temperature and maximum temperature during incineration were measured and the amount of combustion residues and their content of unburned and sintered material were determined. Some ash samples from the combustion of glycerine/pine wood mixes were sent for chemical analysis. The admixture of glycerine in the fuel types generally resulted in pellets with lower durability, higher proportion of fines (fine fraction) and significantly lower bulk density. However, pellets with the same or slightly better durability and the same or lower proportion of fines were obtained by admixture of 1% acidic glycerine to the fuel types. The specific energy consumption for pelleting the fuels containing glycerine was generally lowered, which may have been due to the lubricating properties of the glycerine. The addition of glycerine in all cases resulted in a decrease in pellet bulk density. Reed canary grass had the best durability, least share of fines and highest bulk density, while straw usually had the worst durability, highest share of fines and lowest bulk density. During the storage experiments the pellets took up some water, but never to such a level that the storability was compromised. Pellets with glycerine took up more water, often an increasing amount with increasing glycerine content. Mould was not found in any pellets from any experiment. Durability deteriorated more frequently with higher glycerine content, and thus low durability in the pellets at the outset. The initial moisture content and thus the type of fuel were also of great importance for moisture absorption. Straw contained most moisture from the start and took up the most water. In addition, durability decreased most for straw during storage. During incineration, the main problem with an increasing admixture of glycerine in the fuels was an increasing amount of particles in the flue gas. Calculations of potassium (K) losses with the flue gases, based on analysis of fuels and their ash composition, suggested that this increased sharply with increasing admixture of glycerine. For example, incineration of pine wood with about 1% added glycerine increased the amount of K leaving with the flue gases to the level observed with straw incineration. There seemed to be a close link between the amount of particles in the flue gas and the amount of K released. It is therefore likely that addition of glycerine will cause problems with corrosive deposits in the flues for fuels that normally do not present such problems. However, with fuels that normally give rise to such problems, e.g. straw, it is likely that there will be little obvious difference. Glycerine (acid) containing sulphur can probably prevent particle release and thus lower the risk of corrosion associated with alkali chlorides in the flue gas. The emissions of acrolein and formaldehyde did not increase as might be expected in experiments with glycerine admixture. The levels of acrolein were below the detection limit, and the levels of formaldehyde were very low. Emissions of CO, NOx and SO2 varied widely between experiments in a more or less random way. However, CO emissions may have increased with lower pellet quality. Sulphur dioxide should be produced during incineration of pellets with high sulphur levels. Ash melting point declined in pine wood in particular on addition of glycerine, but also in reed canary grass. The ash melting point for straw was not affected at all by glycerine admixture. The reason why pine wood ash was more sensitive is that the amounts produced from pine are so small that properties of the glycerine ash can have an impact even at small admixture rates. The K content of reed canary grass is so low that very little glycerine with high K content is required before its properties start to become apparent. However, straw has a high initial content of an ash that is already high in K, so the K content of glycerine ash is les obvious even at rather high admixture rates. Calculations using key numbers showed that the ash melting point of reed canary grass in particular should be lowered by the admixture of glycerine, but also that of pine wood to a lesser degree, from an initially high level. Straw ash should scarcely be affected at all. High losses of K in the flue gases can suppress the problem of ash melting point depression in bottom ash. Some of the glycerine types initially had high contents of methanol. It is important that this can be removed, as it constitutes a fire hazard during pelleting and also during transport and handling of the glycerine. High methanol concentrations are also a human health risk, as methanol is toxic. The commercial value of the glycerine is at one of three different levels depending on how it functions when mixed in different fuels. A) If glycerine admixture degrades a good fuel such as wood in terms of the risk of acid deposits in the flue so much that it is similar to straw as a fuel, glycerine value has a negative value. B) On admixture of glycerine in a poor fuel, such as straw, the properties of the glycerine are not evident and its value based on the energy is therefore the same as for the pure fuel. This is currently approximately SEK 0.6-0.9/kg glycerine, which can be compared with the current price of SEK 1/kg for glycerine in biodigestion applications. These factors combined make it difficult to find a profitable use for glycerine as an additive in pelleting or when used as fuel. Glycerine would therefore not be of commercial interest for use in these applications. C) If the glycerine acts as an additive in very small amounts at pelleting, most likely as a lubricant, the value would then be about SEK 3/kg. However, only small amounts would be used in this application, probably just a few hundred tonnes per annum. The advice to the industry is that glycerine should not be used in pelleting as it probably does not work well for this application. If glycerine is incinerated, it should be co-fired with a poor fuel, so as not to impair the properties of the fuel. Glycerine is probably more profitable in applications other than fuel to produce heat. At low addition rates glycerine could probably act as a lubricant during pelleting of certain fuels. However, more research is required to determine whether glycerine can act as a lubricant along with some cheap binding agent in pelleting of fuels

Keywords

biobränsle; glycerin; glycerol; inblandning; tillsatsmedel; pelletering; lagring; eldning; emissioner; nyckeltal; biofuel; glycerine; glycerol; admixture; additive; pelleting; storage; incineration; emissions; key numbers

Published in

Rapport (Institutionen för energi och teknik, SLU)
2011, number: 034
Publisher: Department of Energy and Technology, Swedish University of Agricultural Sciences