Previous PageTable Of ContentsNext Page

Glycerol as a by-product of biodiesel production in Diets for ruminants

Angela Schröder and Karl-Heinz Südekum

Institute of Animal Nutrition, Physiology and Metabolism, University of Kiel, 24098 Kiel, Germany, e-mail:


Glycerol can be derived from the production of biodiesel. It is a glucogenic substance similar to propylene glycol, which has been used effectively to prevent ketosis in high yielding dairy cows. This study was conducted to evaluate the potential of glycerol of three different purities in diets for ruminats. Energy concentrations of glycerols and of glycerol containing diets were determined in vitro and in vivo as related to different types of concentrates, i. e., high in starch versus low in starch. Glycerols at concentrations of up to 10% of total diet dry matter were compared with starch as rapidly fermentable carbohydrate source as related to estimates of ruminal fermentation, microbial biomass production, and nutrient digestibilities in steers at 85% of ad libitum intake. Additionally, we evaluated the physical, chemical and hygienic qualities of concentrate pellets containing glycerol of different purities at different concentrations and stored under different environmental conditions.

Pellet quality, in particular hygienic quality, was positively influenced by glycerol. Estimated energy concentrations from digestibility trials in vivo were 8.3 and 9.5 MJ net energy for lactation/kg of glycerol when glycerol was fed in combination with a high and a low starch concentrate, respectively. From the in vivo data it can be concluded that glycerol of different purities can replace rapidly fermentable starches in diets for ruminants up to concentrations of 10% of diet dry matter without negatively affecting feed and water intake, ruminal nutrient degradation and whole-tract nutrient digestibilities. Rumen microbial biomass production was not different among the diets containing starch or glycerol. The glucose precursor glycerol may improve energy supply to high yielding dairy cows, both before and after calving and thereby have an impact on health and performance during the entire lactation.

KEYWORDS: chemical composition, energy value, ruminal fermentation, starch, pellet quality


In the European Union the turn towards renewable energy sources has increased the production of biodiesel from rapeseed oil (rapeseed oil methyl ester), leaving glycerol as a valuable by-product. Glycerol is a natural, liquid substance of sweet taste which is registered in the European Union as feed additive E 422 (Anonymous, 1995). Lebzien and Aulrich (1993) have reported a high energy concentration (9.5 MJ of net energy for lactation/kg) and glycerol may therefore have benefits to prevent keto-acidosis in the the high yielding dairy cow by increasing the supply of glucose precursors (Johnson, 1955; Fisher et al., 1971; 1973; Sauer et al., 1973). Because data from the United States suggest that 30 to 50% of all dairy cows are affected by subacute ketoacidosis (Hutjens, 1996), means by which energy nutrition of the periparturient cow may be improved are still of special importance.

No legal restrictions as related to animal species or quantity that may be fed to an animal per day are laid upon glycerol (Anonymous, 1995). Therefore, glycerol could become attractive for ruminants and in particular dairy cattle if the amount of the by-product glycerol from biodiesel production exceeds the capacities of the pharmaceutical and chemical industries to process glycerol. Because in the present situation glycerol of high purity is costly, impurer qualities of glycerol should be evaluated as well.

The objectives of our studies were to evaluate the potential of glycerol in diets for ruminants. We studied the chemical composition and energy concentrations as derived from digestibility trials on sheep of glycerol of three different purities. Ruminal nutrient turnover and ruminal microbial biomass in cattle and whole-tract nutrient digestibilities by cattle of diets containing glycerol were also determined. Finally, we investigated the effects on physical, chemical and hygienic variables of concentrate pellets containing glycerol as related to different purities and concentrations of glycerol in the feeds and stored for different durations and under varying environmental conditons.

Chemical composition of glycerol

Glycerols of three different purities (low, medium, high) reflected different stages of the same process of rapeseed oil methyl ester production. The most pronounced variations among purities were the concentrations of water, glycerol, phosphorus and methanol (Table 1).

Table 1. Chemical composition of glycerol as related to purity


Purity of glycerol





Water, %




Composition of the dry matter1, %





Ether extract
























1 Concentrations of cadmium, mercury and arsenic were below the detection limit.

2 not analysed.

ENERGY concentrations OF GLYCEROL

Concentrations of net energy for lactation (NEL) of glycerol as related to purity of glycerol and glycerol content in the dry matter of mixed diets were derived from digestibility trials with wethers. Glycerols were mixed with forage (wilted grass silage) and a high-starch concentrate. The mixed diets were formulated to contain (dry matter basis) 40% forage, 50% concentrate and 10% of pure glycerol irrespective of the purity of the glycerol-containing product. A 100% forage diet and a glycerol-free forage:concentrate (40:60) diet served as controls. In addition, the glycerol of the highest purity (100% glycerol) was included in diets containing 40% forage and 5, 10, 15 or 20% glycerol. The high-starch concentrate or a low-starch concentrate made up the balance of these diets (55 to 40% of the dry matter). The NEL concentrations were higher when glycerol was fed with the low-starch concentrate than when fed with the high-starch concentrate. Estimated energy concentrations for glycerol derived from diets containing the low-starch concentrate and 10, 15 or 20% glycerol in the dry matter were similar. The mean value was 9.7 MJ NEL/kg glycerol, which is very close to that reported by Lebzien and Aulrich (1993) from a trial with dairy cows. When fed with the low-starch concentrate, either no effect or positive effects of glycerol on nutrient digestibilities (organic matter, starch, cell-wall components) were observed. When derived from diets containing the high-starch concentrate as opposed to the low-starch concentrate, glycerol of different purities and at different dietary inclusion levels (10, 15 or 20%) had markedly lower NEL concentrations (8.0 to 8.5 MJ/kg). Similarly, digestibilities of cell-wall components were depressed yet without reducing organic matter digestibilities. In vitro estimates of NEL concentrations based on gas production of feeds incubated with ruminal fluid were slightly lower than those obtained in vivo and were similar for the low- and the high-starch diets.

Effects of glycerol on Ruminal turnover, microbial biomass in the rumen and ON WHOLE-TRACT NUTRIENT digestibilities

Four ruminally cannulated steers were used in a 4 × 4 Latin square design. They were fed on mixed diets (40:60 forage:concentrate, dry matter basis). The concentrate pellets were isonitrogenous and contained no glycerol or 15% glycerol from glycerols of different purities (Table 1). Although methanol concentration of the glycerol of low purity was marked, only negligible quantities of methanol were detected in the pelleted concentrates. Daily dry matter intake averaged 13.4 kg. With the diets containing glycerol, the animals consumed 1.1 kg glycerol and 1.4 kg starch per day, whereas the daily starch intake with the glycerol-free diet was 2.1 kg. Total tract digestibilities of organic matter, cell-wall (neutral detergent fibre) and starch were similar for all dietary treatments (mean value, 72.1, 65.1 and 98.3%, respectively). Despite the high percentage of concentrate in the diet and irrespective of dietary treatment, postprandial pH values in ruminal fluid were always greater than 6.2. The postprandial decline in pH was more pronounced when the diets contained glycerol, indicating that ruminal degradation of glycerol was faster than that of wheat starch. Feeding glycerol resulted in a slight shift towards a reduced ratio of acetic acid to propionic acid, which was significant (control versus glycerol-containig diets, P < 0.05) for most of the single time points that were measured at hourly intervals over the 24-h cycle. Although butyric acid concentrations in the ruminal fluid of the control diet were almost constant throughout the day, the values increased markedly postprandially and peaked significantly (control versus glycerol-containig diets, P < 0.05) at 3 hours postfeeding for all diets containing glycerol. The increase in butyric acid concentrations may be benificial for several reasons (van Houtert, 1993). Butyric acid is metabolised to ketone bodies by rumen and omasal epithelia, which has been proposed as a mechanism to provide the rumen epithelium with most of its energy requirements. Because butyric acid can inhibit cell division, conversion to β-hydroxybutyrate can aid as a means of detoxification. Additionally, ketone bodies provide extra-hepatic tissues with an additional energy source and β-hydroxybutyrate may play an important role in metabolic regulation and control of feed intake in ruminants. Lactic acid was only detected up to 4 hours postfeeding, albeit with a peak value of 47 mmol/L for the diet containing the glycerol of low purity. However, only the concentrations at 1 hour postfeeding were different (P < 0.05) between the control and the glycerol-containig diets. Bacterial dry matter concentration in the rumen fluid was estimated from optical density of the rumen fluid at a wavelength of 600 nm (Wejdemar, 1993). There was a slight decrease of bacterial dry matter concentration in the diets with glycerol, but because of large within diet variation the observed differences were statistically not significant. Daily water intake was slightly higher when cattle consumed diets containing the glycerol of low and medium purity. Rumen fill also was slightly higher with the diets containing glycerol. The proportion of bailable liquids of total ruminal contents was 22, 25, 29 and 31% for the diets containing no glycerol or the glycerol of low, medium and high purity. Obviously, glycerol had an impact on ruminal water turnover. Estimates of ruminal in vivo fermentation of nutrients were based on manual emptying of the rumens (Robinson et al., 1987). The ruminal in vivo fermentation of fibre components (mean value, 51.7%/day) was not impaired when glycerol was substituted for starch in the concentrate portion of the diets.

Effects of glycerol on quality OF PELLETED CONCENTRateS

Concentrates were formulated that consisted (% of dry matter) of glycerol (0, 5, 10, 15), wheat and soybean meal (30 to 10 and 10 to 15, respectively, depending on glycerol concentration), and of 12 rapeseed meal, 20 dried beet pulp 14 wheat bran, 10 maize and 4 vitamin-mineral mix. Prior to pelleting, the ingredients were ground through a sieve with a 3-mm screen. The pellet press consisted of three rotating rollers and a static die. Because of varying concentrations and different purities of the glycerols in the concentrates, a sensitive adjustment of the pelleting process was necessary to achive pellets of the best possible quality. Therefore the gap-size between the roller and the die as well as the distance between the pellet knife and the die face were varied based on the experience of the operator.

Treatments were arranged as a 10 × 5 × 3 factorial design, with ten concentrate types, five levels of environments and three replicates of each combination of factors. The ten concentrate types resulted from one concentrate without glycerol (control) and three glycerol concentrations (5, 10, 15% of dry matter) × three purities of glycerol (Table 1). Environments were defined by an unstored control and two storage environments (good: 15°C and 60% relative humidity and bad: 20°C and 70% relative humidity), both combined with two storage durations, namely four and eight weeks.

Storage environment and storage duration did not affect the chemical composition of the concentrates. Only the dry matter concentrations differed, which may reflect the hygroscopic properties of glycerol. Within each purity – except for the concentrates with the glycerol of high purity – the dry matter content of unstored concentrates increased with increasing concentration of glycerol. Contrarily, lower dry matter contents, i. e., higher moisture concentrations, were obtained with higher concentrations of glycerol for all concentrates that were stored for four or eight weeks. Effects of storage and concentrate types on dry matter content, as well as the interaction between these effects, were significant. In spite of the increasing moisture the hygienic quality of the pellets was excellent. The majority of values of ergosterol concentrations, which were used as a chemical indicator of fungal biomass, were considerably lower than the 'preliminary normal ranges' (Müller and Schwadorf, 1988; 1990). These values, however, must be treated with caution and the presence of mycotoxins can not be excluded from one single variable like ergosterol (Müller and Schwadorf, 1988). Only the storage of the glycerol-free concentrate under bad environmental conditions for eight weeks resulted in an increased ergosterol concentration (12.9 mg/kg of dry matter). Although fungal biomass obviously increased markedly in this concentrate, the chemical composition remained virtually unchanged as compared with the unstored treatments. The results indicate that glycerol of different purities and even at concentrations as low as 5% of dry matter had a preserving effect on pelleted concentrates.

The most pronounced positive effects of glycerol on hardness of pellets were observed for the concentrate types with 10% glycerol and for those of low purity across all storage environments. The latter can be attributed to the high water input by the glycerol of low purity. Overall, the concentrate without glycerol (control) had the highest abrasion and only medium values for pellet hardness. This may indicate a positive effect on these variables of glycerol per se. Storage per se (unstored versus stored) increased the hardness of the pellets, which could be due to the moisture content, and thereby affecting binding phenomena between particles, whereas duration (four weeks versus eight weeks) did not affect pellet hardness. These effects were closely combined with effects on abrasion of fines from the pellets: The harder the pellets, the lower the abrasion. However, although abrasion was low, in some cases the pellets with 15% pure glycerol were elastical and fir-coned.


Glycerol of different purities can be included in mixed diets for ruminants up to 10% of the dry matter as a substitute for rapidly fermentable starch sources, e. g., wheat or tapioca, without negatively affecting ruminal environment, ruminal nutrient turnover and whole-tract digestibilities of organic matter constituents. Mean energy concentrations of glycerol derived from diets fed to sheep and containing a low-starch or a high-starch concentrate were 9.7 and 8.3 MJ NEL/kg of glycerol, respectively. When fed with a low-starch concentrate, pure glycerol at dietary inclusion levels up to 20% had no effect or positive effects on nutrient digestibilities. When included in diets containing high-starch concentrates, glycerol reduced cell-wall digestibilities but had no obvious effect on whole-tract organic matter digestibilities. Chemical as well as physical pellet quality variables were not affected greatly. The preserving effect on concentrate pellets of the glucogenic substance glycerol must be emphasized.

The results of these studies suggest that the glucose precursor glycerol is an excellent feed constituent, even when included in an impure form as derived from biodiesel production. Glycerol may serve as an ingredient both of pelleted concentrates or of total mixed rations. In pelleted concentrates, the contribution to the hygienic quality of the feedstuff might be of special interest. Economic assessment will be decisive of a wider use of glycerol as a dietary ingredient for ruminants.


This study was supported in part by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation, SCHR 551/1-1), Bonn, Germany, and the Union zur Förderung von Oel- und Proteinpflanzen (UFOP e. V., Bonn, Germany, 524/951 and 524/964). We thank C. Lewin and A. Ombabi for animal care and for their unremitting help with preparing the concentrate pellets. W. Kühl, U. Lies, M. Paschke-Beese, R. Schwer and A. Weßmann are gratefully acknowledged for skilled analytical assistance. Dr. H. Steingaß is acknowledged for conducting the in vitro measurements.


1. Anonymous, 1995. Futtermittelrecht mit einschlägigen Bestimmungen, Bundesgesetzen, Verordnungen, Erlassen und Recht der Europäischen Union, 2nd edition, supplement 6, F-4, appendix I, 12a, Blackwell Wissenschafts-Verlag, Berlin.

2. Fisher, L.J., Erfle, J.D., Lodge, G.A. and Sauer, F.D., 1973. Effects of propylene glycol or glycerol supplementation of the diet of dairy cows on feed intake, milk yield and composition, and incidence of ketosis. Canadian Journal of Animal Science 53, 289-296.

3. Fisher, L.J., Erfle, J.D. and Sauer, F.D., 1971. Preliminary evaluation of the addition of glucogenic materials to the rations of lactating cows. Canadian Journal of Animal Science 51, 721-727.

4. Hutjens, M.F., 1996. Practical approaches to feeding the high producing cow. Animal Feed Science and Technology 59, 199-206.

5. Johnson, R.B., 1955. The treatment of ketosis with glycerol and propylene glycol. Cornell Veterinarian 44, 6-21.

6. Lebzien, P. and Aulrich, K., 1993. Zum Einfluss von Glycerin auf die Rohnährstoffverdaulichkeit und einige Pansenparameter bei Milchkühen. VDLUFA-Schriftenreihe 37, 361-364.

7. Müller, H.M. and Schwadorf, K., 1988. Ergosterin - ein Parameter zur quantitativen Bestimmung des Pilzbesatzes von Futtermitteln. Kraftfutter/Feed Magazine 71, 174-178.

8. Müller, H.M. and Schwadorf, K., 1990. Ergosterin als Mass für das Pilzwachstum in Futtermitteln 2. Mitteilung: Ergosteringehalt von Mischfutterkomponenten und Mischfuttern. Archives of Animal Nutrition 40, 385-395.

9. Robinson, P.H., Tamminga, S. and van Vuuren, A.M., 1987. Influence of declining level of feed intake and varying the proportion of starch in the concentrate on rumen ingesta quantity, composition and kinetics of ingesta turnover in dairy cows. Livestock Production Science 17, 37-62.

10. Sauer, F.D., Erfle, J.D. and Fisher, L.J., 1973. Propylene glycol and glycerol as a feed additive for lactating dairy cows: An evaluation of blood metabolite parameters. Canadian Journal of Animal Science 53, 265-271.

11. van Houtert, M.F.J., 1993. The production and metabolism of volatile fatty acids by ruminants fed roughages: A review. Animal Feed Science and Technology 43, 189-225.

12. Wejdemar, K., 1993. Effects of concentrations and combinations of carbohydrates on growth of four species of rumen bacteria in rumen fluid. Swedish Journal of Agricultural Research 23, 141-148.

Previous PageTop Of PageNext Page