Crude glycerin as an alternative energy feedstuff for dairy cows

Crude glycerin as an alternative energy feedstuff for dairy cows

Animal Feed Science and Technology 183 (2013) 116–123 Contents lists available at SciVerse ScienceDirect Animal Feed Science and Technology journal ...

529KB Sizes 0 Downloads 7 Views

Animal Feed Science and Technology 183 (2013) 116–123

Contents lists available at SciVerse ScienceDirect

Animal Feed Science and Technology journal homepage:

Crude glycerin as an alternative energy feedstuff for dairy cows Cassio Andre Wilbert a,∗ , Ênio Rosa Prates b , Júlio Otávio Jardim Barcellos c , Jorge Schafhäuser d a Post-Graduation Program in Animal Science, Department of Animal Science, School of Agronomy, Federal University of Rio Grande do Sul (UFRGS), Rio Grande do Sul 91540-000, Brazil b Department of Animal Science, School of Agronomy – UFRGS, Rio Grande do Sul 91540-000, Brazil c Department of Animal Science, School of Agronomy – UFRGS, Rio Grande do Sul 91540-000, Brazil d Embrapa - Center for Temperate Climate Agricultural Research, Rio Grande do Sul 96010-971, Brazil

a r t i c l e

i n f o

Article history: Received 4 September 2012 Received in revised form 3 May 2013 Accepted 6 May 2013

Keywords: Digestibility Glycemia Intake Milk composition Milk yield Non-esterified fatty acids

a b s t r a c t The intent of this study was to demonstrate the potential of biodiesel derived crude glycerin as a good alternative for an energy feedstuff in dairy cow diets. Eight multiparous Jersey cows with 85 ± 20 days in milk were used. The following treatments were evaluated: 0, 40, 80, and 120 g crude glycerin (containing 814.4 g glycerol/kg) dietary inclusion/kg dry matter. The experimental diets contained equal protein and energy levels. A replicated Latin square experimental design was applied. Crude glycerin intake did not influence average daily milk yield, energy corrected milk yield, or crude fat, lactose, and total milk solids average daily yield and concentration. Milk protein concentration was higher when 120 g crude glycerin was included in the diet when compared with the control group (37.2 vs. 36.1 g/kg, respectively, P<0.05). Crude glycerin intake also increased average daily crude protein yield at the dietary inclusion levels of 80 g/kg (0.77 vs. 0.71 kg by the control group, P<0.05). None of the evaluated treatments influenced dry matter or organic matter intake. Dry matter, organic matter, and neutral detergent fiber digestibility, and serum concentrations of non-esterified fatty acids and urea were not influenced by crude glycerin intake. Glycemia showed a quadratic response to crude glycerin intake (P<0.05), and was lower at the intermediate levels of dietary inclusion (40 and 80 g/kg). Crude glycerin is a good alternative energy feedstuff for dairy cows at a dietary inclusion levels of up to 120 g/kg dry matter in partial substitution of ground corn in dairy cow rations, and may even improve milk protein concentration. © 2013 Elsevier B.V. All rights reserved.

1. Introduction The inclusion of 20 mL of biodiesel/L fossil fuel diesel is mandatory in Brazil, since 2008 (ACT n. 11.097, as of January 13, 2005). In 2010, Resolution CNPE n. 6/2009 allowed the inclusion of up to 50 mL/L. This has resulted in 2011 in the production of more than 2.64 billion liters of biodiesel in Brazil, whereas between March and December, 2005, only 736,100 L were

Abbreviations: CGL, crude glycerin; CP, crude protein; DM, dry matter; DMD, dry matter digestibility; DMI, dry matter intake; ECM, energy corrected milk; EE, ether extract; aNDFom, neutral detergent fiber with a heat stable amylase and expressed exclusive of residual ash; NDFD, neutral detergent fiber digestibility; NEl, net energy of lactation; OM, organic matter; OMD, organic matter digestibility; OMI, organic matter intake; RDP, rumen degradable protein; VFA, volatile fatty acids. ∗ Corresponding author. Tel.: +55 051 3339 5126; fax: +55 051 3308 6048. E-mail address: [email protected] (C.A. Wilbert). 0377-8401/$ – see front matter © 2013 Elsevier B.V. All rights reserved.

C.A. Wilbert et al. / Animal Feed Science and Technology 183 (2013) 116–123


produced (ANP, 2012). This huge increase in the production of biodiesel is not restricted to Brazil. The estimated global biodiesel production for 2019 is of about 41 billion liters (OECD-FAO, 2010). Although this growth in productions creates to some problems, it may also create opportunities. It estimated that each liter of biodiesel produced generates about 100 mL crude glycerin (CGL) (Dasari et al., 2005), which contains variable glycerol content. A large surplus of CGL has been recently produced, and a nobler destiny than routine burning and disposal in rivers and industrial dump, need to be found in order to reduce environmental pollution and improve the economic return of the biodiesel producers. However, complex and costly processes are required to obtain the degree of purity required for the traditional applications of glycerin in food, drug, cosmetic and tobacco industries cosmetics, drugs, and cleaning products (Pachauri and He, 2006), because it contains, several impurities including residual methanol, NaOH, carry-over fat/oil, some esters, and low amounts of sulfur compounds, proteins, and minerals (Celik et al., 2008). Glycerol, the main component of CGL, has high energy content, which is approximately the same as that of corn starch (Donkin and Doane, 2007), and can potentially be used for animal feeding. In general CGL was used as a supplement, increasing energy intake, in beef cattle diets, and nutrient utilization was not reported for those studies (Schröder and Südekum, 1999; Mach et al., 2009; Wang et al., 2009; Lage et al., 2010). Therefore, further studies are required to evaluate the use of CGL as an alternative energy feedstuff to replace corn grain, the main energy feedstuff in dairy diets. The objective of the present study was to demonstrate that CGL derived from a biodiesel production plant may be an interesting alternative energy feedstuff for dairy cows in the middle third of lactation fed iso-energy and iso-protein diets, thereby upgrading its status as residue to important by-product. In addition, the specific aim was to determine maximal CGL dietary inclusion level, up to which no negative effects on production and feed utilization are observed. 2. Materials and methods 2.1. Animals and location The field experiment was carried out between September and December, 2009, at the Dairy Cattle System sector (SISPEL) of the experimental farm Terras Baixas (ETB) of the Temperate Climate Agriculture Research Center of Empresa Brasileira de Pesquisa Agropecuária (EMBRAPA), located at the municipality of Capão do Leão (31◦ 52 20 south latitude and 52◦ 21 24 west longitude, mean altitude of seven meters above sea level), Rio Grande do Sul, Brazil. Milk component analyses were carried out at the Milk Quality Laboratory (LABLEITE) located at that experimental station. Chemical analyses were performed at the Animal Nutrition Laboratory and the biochemical analyses in the Veterinary Clinical Analyses Laboratory, both belonging to Universidade Federal do Rio Grande do Sul (UFRGS). The work was carried out in accordance with EU Directive 2010/63/EU. Eight purebred Jersey cows, of 2–8 parities, with average body weight of 421 ± 39 kg and 85 ± 20 days of lactation, were selected from the SISPEL herd. Average daily milk yield before the beginning of the experiment was 20.5 ± 2.7 kg. Cows were individually housed in free stall, and were separated by an electrical fence. All pens were equipped with a drinker, dispensing drinkable water, and sand litter. 2.2. Treatments The following treatments were evaluated: Control – no CGL inclusion in the diet; G4 – 40 g CGL/kg of dietary dry matter (DM); G8 – 80 g CGL/kg dietary DM; G12 – 120 g CGL/kg dietary DM. CGL inclusion levels are described on dry matter basis. The CGL tested contained 814.4 g glycerol/kg DM, 860.1 g of DM/kg, 1.1 g crude protein (CP)/kg of DM, 14.29 MJ of gross energy (GE)/kg of DM and less than 50 ppm of methanol. Glycerol was considered as completely digestible during diet calculation. Metabolizable energy (ME) was estimated as: ME = (1.01 × DE – 0.45), and net energy of lactation (NEl) was estimated as: NEl = (0.703 × ME) – 0.19. In both equations, units are expressed as calories/kg, according to the NRC (2001). Energy content, NEl, was then transformed in MJ/kg DM, and CGL NEl was estimated as 8.20 MJ/kg of DM. Metabolizable energy (ME) of the other feedstuffs was also estimated based on their chemical composition, also according to the NRC (2001). When feedstuffs contained less than 30 g ether extract (EE)/kg of DM, ME = DE*1.01–0.45, when EE/kg was higher than 30 g/kg of DM, ME = DE*1.01 − 0.45 + 0.0046*(EE – 3). The same equation that was used to estimate the NEl of CGL was used to estimate the NEl of the other feedstuffs with less than 30 g of EE/kg of DM. For feedstuffs with more than 30 g of EE/kg of DM, the following equation (NRC, 2001) was applied: NEl = (0.703*ME) − 0.19 + ((0.097*ME + 0.19)/97)*(EE – 3). Feedstuff digestible energy (DE) values were obtained from the NRC tables (2001), as they were used only to supply equal energy density to the experimental diets. As mentioned before, because the models were built using as unit calories/kg, results were transformed in MJ/kg. The experimental diets were formulated with the aid of the software program Spartan® (Michigan State University). The diets contained equal energy and protein levels and were formulated to supply the requirements of cows producing 20 kg of milk/day containing 48 g of fat/kg, according to the NRC (2001). The NEl estimated based on the chemical composition of the CGL was very similar to the energy density obtained with ground corn. Therefore, CGL entered in diet composition almost exclusively in replacement of corn grain.


C.A. Wilbert et al. / Animal Feed Science and Technology 183 (2013) 116–123

Table 1 Composition of diets consumed. Composition Ingredient (g/kg of DM) Sorghum silage Alfalfa hay Corn grain, ground Soybean meal Crude glycerin Mineral and vitamin premixa Sodium bicarbonate Chemical composition (g/kg DM) DM (g/kg) OM CP aNDFomb NEl (MJ/kg of DM)





274.1 245.2 291.9 167.8 – 20.0 1.0

275.1 246.1 249.2 168.5 40.1 20.0 1.0

271.7 243 207.7 177.4 79.2 20.0 1.0

271.4 242.7 158.6 187.7 118.6 20.0 1.0

726.1 924.7 165.7 338.0 6.53

724.8 922.4 162.5 335.3 6.53

726.0 920.3 162.5 329.3 6.53

725.5 917.6 163.1 326.1 6.53

Control = without crude glycerin; G4 = 40 g of crude glycerin/kg of DM; G8 = 80 g of crude glycerin/kg of DM; G12 = 120 g of crude glycerin/kg of DM. a Contained (g/kg): Ca (minimum) 220 g; P (minimum) 95 g; Mg (minimum) 12 g; Na (minimum) 60 g; S (minimum) 12 g; Vit. A (minimum) 120,000 UI; Vit. D3 (minimum) 30,000 UI; Vit. E (minimum) 750 mg; Se 20 mg; Zn 3000 mg; F (maximum) 950 mg. b Neutral detergent fiber using a heat stable amylase and corrected for ash.

Table 1 presents the ingredient and chemical composition of the experimental diets. 2.3. Experimental management Approximately 14 days before the beginning of the experiment, cows were housed in stalls and adapted to the experimental control diet, during which milk yield and feed intake were measured. In every experimental period, were used 10 days to adaptation of the animals to the diets, and seven days of data collection. The experimental diets were formulated for a minimal intake of 500 g of roughage/kg of total dry matter intake (DMI). Allowance was adjusted whenever a minimum of 10% feed residue was measured in order to ensure ad libitum feed intake. Cows were fed after each milking (8:00 and 17:30 h), and concentrated ration and CGL were divided in three meals by supplying one of the meals at 12:00 pm. The concentrated ration was supplied first and mixed with CGL in the trough. Roughage was supplied only after the concentrate ration was totally consumed, thereby ensuring total CGL intake. In the morning, before milking, feed residues were weighed and, if required, allowance was adjusted. During the seven days collection period, fecal production was estimated by supplying 5 g of chromium sesquioxide (Cr2 O3 ) twice daily (10 g/h/d) mixed with approximately 100 g of concentrate after each milking. Fecal samples were collected directly from the rectum twice daily, immediately after milking and before cows were fed, and were placed in plastic bags, duly identified, and stored in a freezer (−8 ◦ C). At the end of each collection period, fecal samples were thawed and an aliquot of approximately 100 per collection shift was taken and homogenized to obtain a sample pooled by period, animal, and treatment. Immediately after the morning milking, on days 12 and 16 of the experimental period, blood was collected by venipuncture of the jugular vein. Samples were stored in tubes without anticoagulant until the clot was formed, and then centrifuged at 8000 RPM for 10 min for serum separation. The serum was separated and transferred to Eppendorf tubes, which were stored in a freezer (−8 ◦ C) for subsequent determination of non-esterified fatty acids (NEFA) and urea. At the same time, glucose blood concentration was also determined using the commercial kit Accu-check® (Abbot Laboratories S.A.). 2.4. Measurements Individual milk yield was automatically measured at both daily milkings (7:00 and 17:00) during the entire experimental period, and average milk yield for each period was calculated based milk produced between days 11 and 17 of each period. Average daily milk yield was corrected for energy (ECM), according to the following equation (Sjaunja et al., 1990): ECM (kg/d) = kg milk × ((383Fat% + 242Protein% + 165.4Lactose% + 20.7)/3140). Milk samples of two consecutive milkings (morning and afternoon) of all experimental animals were collected on days 11, 14, and 17 of each period. After milk yield was measured, the milk present in the collecting tubes was homogenized and an aliquot was taken. Samples were placed in tubes, to which bronopol (2-bromo.2-nitro-1.3-propanediol) was added, refrigerated, and submitted to the laboratory in up to 24 h. Milk components (fat, crude protein – CP, lactose, and total solids) were determined by infrared spectroscopy, according to the AOAC (1996, method 972.16). Based on data from three collections per period, the average of each milk component was calculated for each experimental period. The following equation was used to determine the concentration of each milk component: Component (g/kg) = (concentration obtained in morning milking × morning milk yield relative to total yield) + (concentration obtained in evening milking × evening milk yield relative to total yield).

C.A. Wilbert et al. / Animal Feed Science and Technology 183 (2013) 116–123


Table 2 Effects of crude glycerin intake on performance (milk yield and ECM, kg/d), milk composition (g/kg) and milk components yield (kg/d). Item Milk yield ECM Milk fat g/kg kg/d Milk protein g/kg kg/d Lactose g/kg kg/d Total solids g/kg kg/d




19.9 19.4

G4 19.8 19.3

G8 20.9 20.1

G12 19.6 18.5

0.518 0.542

0.16 0.29

37.7 0.74

36.7 0.73

35.5 0.74

34.0 0.67

1.062 0.026

0.28 0.36

36.1b 0.71b

36.2ab 0.72ab

36.9ab 0.77a

37.2a 0.73ab

0.330 0.019

0.01 0.05

45.0 0.90

45.4 0.91

44.9 0.94

44.8 0.88

0.425 0.028

0.53 0.25

129 2.54

128 2.55

128 2.67

126 2.47

1.386 0.070

0.42 0.20

a,b Means within a line with different superscript letters differ (P<0,05). Control = without crude glycerin; G4 = 40 g of crude glycerin/kg of DM; G8 = 80 g of crude glycerin/kg of DM; G12 = 120 g of crude glycerin/kg of DM ECM – energy corrected milk = kg milk × ((383Fat% + 242Protein% + 165.4Lactose% + 20.7)/3140) (Sjaunja et al., 1990).

Daily net energy for lactation was calculated as follows (NRC, 2001): NEl (Mcal) = (0.0929*fat (g/kg) + 0.0547*CP (g/kg) + 0.0395*Lactose (g/kg))*average milk yield (kg). The obtained values were then transformed in MJ/kg. Non-esterified fatty acids (NEFA) analyses were performed using commercial kits (Randox® ) and readings were made in an automatic analyzer (Metrolab D-1600® ). Urea analyses were carried out using the commercial kit Ureia 500® (Doles S.A.), and readings were made in a spectrophotometer. Neutral detergent fiber (aNDFom) was analyzed according to Van Soest et al. (1991) using a heat stable ␣-amylase, without the use of sodium sulfite and was corrected for ash. The DM was analyzed according to Easley et al. (1965), and organic matter (OM), CP and EE according to the AOAC (1996, methods 942.05, 954.05 and 920.39, respectively). 2.5. Experimental design and statistical analysis A replicated (4 × 4) Latin Square experimental design, with four treatments and four periods, was applied. Animals were considered experimental units. Every experimental period comprised 10 days of adaptation to the treatments and seven days of evaluation of maximal intake and apparent digestibility. Data were submitted to analysis of variance (ANOVA) using the GLM procedure of SAS statistical package, version 9.0 (SAS, 2002). The following causes of variation were evaluated: Treatment, period, Latin square, cow within Latin square and the interaction Latin square*treatment. Results were considered significant when P<0.05, and a trend when 0.05 < P<0.1. When an effect of treatment was detected with ANOVA, regression analyses (linear, quadratic, and cubic) were performed. The CGL inclusion levels were considered as continuous variables. If the results of the regression analyses were not statistically significant, means were compared by the test of Tukey at 5% significance level. The model applied for ANOVA was: Yijkl =  + Ti + LSj + Ck (LSj ) + Pl + (LST)ij + Eijkl . Where: Yijkl = mean value obtained for each observation;  = general mean of the variable in the experiment; Ti = effect of treatment i; LSj = effect of the Latin square i; Ck (LSj ) = effect of cow k within Latin square j; Pl = effect of period l; (LST)ij = interaction between Latin square j and treatment i; Eijkl = experimental error. And for the analysis of regression of each parameter, the following model was applied: Yij =  ± ␤1Pj + ␤2Pj 2 + ␤3Pj 3 . Where: Yij = ith observation associated to the jth inclusion level of crude glycerin; ␮ = general mean of the variable in the experiment; ␤1Pj = regressor associated with the linear effect of the inclusion of crude glycerin; ␤2Pj 2 = regressor associated with the quadratic effect of the inclusion of crude glycerin; ␤3Pj 3 = regressor associated with the cubic effect of the inclusion of crude glycerin. 3. Results There was no influence of crude glycerin (CGL) on milk yield (kg/d), energy corrected milk yield (ECM, kg/d), concentration (g/L) or average daily yield (kg/d) of fat, lactose or total solids (Table 2). Milk crude protein (CP, g/L) was increased by CGL inclusion level (Table 2; P<0.05), which also tended (P=0.0540) to affect average daily CP yield (kg). Control cows produced milk with 3.05% less CP than those submitted to G12 (Table 2; P<0.05). In G8, cows produced, in average, 8.45% more CP/day than Control cows (Table 2; P<0.05), due to the fact that milk yield of G8 cows was 1.03 kg/d higher than the average daily milk yield of those that were not fed CGL (20.90 vs. 19.87 kg/d), associated to a CP concentration 2.22% higher in G8 than in Control, although no differences were detected between these treatments for milk yield and milk CP content (Table 2). The inclusion of CGL did not influence dry matter (DMI) or OM (OMI) intake, either in absolute values (kg/d) or relative to body weight (g/kg body weight) (Table 3).


C.A. Wilbert et al. / Animal Feed Science and Technology 183 (2013) 116–123

Table 3 Effects of crude glycerin intake on dry matter intake (DMI, kg/d and g/kg of body weight), organic matter intake (OMI, kg/d and g/kg of body weight). Item DMI kg/d g/kg of BW OMI kg/d g/kg of BW







19.1 47.1

18.5 45.3

18.5 45.6

18.5 45.0

0.407 1.068

0.75 0.44

17.7 43.8

17.6 42.0

17.1 42.1

17.1 41.6

0.379 0.996

0.66 0.36

Control = without crude glycerin; G4 = 40 g of crude glycerin/kg of DM; G8 = 80 g of crude glycerin/kg of DM; G12 = 120 g of crude glycerin/kg of DM. DMI = dry matter intake; OMI = organic matter intake; BW = body weight. Table 4 Effects of crude glycerin intake on dry matter (DMD), organic matter (OMD) and neutral detergent fiber digestibility (NDFD) coefficients. Item








0.65 0.68 0.46

0.65 0.68 0.47

0.65 0.68 0.42

0.66 0.69 0.45

0.006 0.005 0.010

0.93 0.95 0.41

Control = without crude glycerin; G4 = 40 g of crude glycerin/kg of DM; G8 = 80 g of crude glycerin/kg of DM; G12 = 120 g of crude glycerin/kg of DM. DMD = dry matter digestibility; OMD = organic matter digestibility; NDFD = neutral detergent fiber digestibility. Table 5 Effects of crude glycerin intake on glucose (mg/dL), non-esterified fatty acids (NEFA, mmol/L) and blood urea (mg/dL). Item a

Glucose NEFA Blood urea







70.8 0.21 51.2

66.9 0.17 47.6

69.3 0.18 44.3

72.0 0.19 48.4

0.674 0.009 1.199

0.04 0.38 0.23

Control = without crude glycerin; G4 = 40 g of crude glycerin/kg of DM; G8 = 80 g of crude glycerin/kg of DM; G12 = 120 g of crude glycerin/kg of DM NEFA = non-esterified fatty acids. a Linear (P=0.2946); Quadratic (P=0.0104); Cubic (P=0.2652).

Glucose (mg/dL)


80 70 60 50 0




Crude glycerin (g/kg of DM) Fig. 1. Effect of crude glycerin inclusion on the serum glucose: Glucose (mg/dL) = 70.566–1.1086*crude glycerin + 0.1045*crude glycerin2 ; r2 = 0.2288.

The DM, OM, and aNDFom digestibility (respectively, DMD, OMD and NDFD) digestibility coefficients were not influenced by the different CGL levels (Table 4). Crude glycerin levels had a quadratic influence on blood glucose (Table 5; P<0.05; Fig. 1; P<0.05), with an initial reduction in G4 followed by an increase in response to CGL intake. No differences were observed in the blood concentrations of non-esterified fatty acids (NEFA) or urea (Table 5). 4. Discussion 4.1. Milk yield As shown in Table 1, the diets offered to the cows contained equal energy and protein levels. Therefore, differences in milk yield would rather result from differences in feed intake or feed digestibility, influencing the utilization of the consumed nutrients, which was not observed in the present study (Tables 3 and 4). There are few studies in literature evaluating high glycerol or CGL inclusion levels (higher than 10%) for dairy cows in the middle third of lactation. One of these studies was carried out by Donkin et al. (2009), who tested four glycerol inclusion levels (0, 50, 100 e 150 g/kg of DM), and did not detect any effects on average daily milk yield. Therefore, as there was no influence of the treatments on average daily milk yield or on average milk fat content, no differences were observed in ECM (Table 2).

C.A. Wilbert et al. / Animal Feed Science and Technology 183 (2013) 116–123


4.2. Milk composition In order to understand the results obtained for milk composition, it is essential to understand the process of glycerol fermentation. Krehbiel (2008) cited three routes have been described for the orally ingested glycerol: passage (13%), fermentation (44%) and absorption (43%). Although the first in vitro studies showed an almost complete fermentation of glycerol, generating propionate, lactate, succinate, and acetate, the main components generated in vivo are propionate and particularly butyrate (Krehbiel, 2008). Carvalho et al. (2011) detected important changes in the fermentation pattern with the inclusion of 110 g of glycerol/kg of DM in the diet of dairy cows during the dry period: there was an increase in the proportion of propionate (22.7% in the control group vs. 28.6% in the glycerol group) and of butyrate (11.5% in the control group vs. 15.3% in the glycerol group) at the expense of acetate (61.4% in the control group vs. 51.5% in the glycerol group. Therefore, when only the percentage of each volatile fatty acid (VFA) is considered, lower milk fat contents could be expected as CGL dietary inclusion level increased due to a dilution effect, as there should be higher availability of lactose precursor (propionate), which is the main osmotic component, accounting for milk volume. This dilution effect was observed by Chung et al. (2007). In order to understand the reason why CGL intake did not influence milk fat content or yield, it must be considered that its maximum inclusion represents only 120 g/kg of total DM, as well as the fermentation characteristics of starch, which is the main rapidly-fermentable carbohydrate of corn. Zeoula et al. (1999) observed a concentration of 793.2 g of starch/kg of DM in Brazilian corn samples, which digestibility coefficient is higher than 0.90 and which also generates essentially propionate (Van Soest, 1994). Considering milk CP content obtained in each treatment, the initial hypothesis is that average daily CP yield had the same behavior, and therefore, the difference between Control and G12 would persist, which was not the case. Although there was no effect of treatment on milk yield, G8 presented 20.9 kg average daily milk yield, whereas Control, 19.87 kg. Therefore, when this parameter is associated with average daily CP content ((36.1 g/kg in Control and 36.9 g/kg in G8) for the calculation of average daily CP production, the higher values of G8 (0.77 vs. 0.71 kg in Control, Table 2, P<0.05) are justified. Increasing blood insulin levels have a positive effect on milk protein synthesis independently of additional amino acid supply, showing the ability of the mammary gland to capture amino acid from sources other than the diet (Mackle et al., 2000). However, as there were no differences in glycemic levels between Control and G12 cows that could possibly influence blood insulin levels (Table 2), this hypothesis can not explain the results observed in the present experiment. Chung et al. (2007) and DeFrain et al. (2004) did not observed effect of glycerin intake over the insulin levels. The study of Mackle et al. (2000) also does not validate the hypothesis that the higher CP concentration obtained in G12 is due to a higher dietary inclusion of soybean meal. In addition, the estimated intake of lysine and methionine (the estimative were made using the NRC, 2001) were not different between the Control and G12 (control: 162 g/h/d of lysine and 52 g/h/d of methionine; G12: 163 g/h/d of lysine and 49 g/h/d of methionine). Another factor that can be considered is that while glycerol is a 3-carbon compound, starch is a 6-carbon probably resulting in a different amino acid profile reaching at the small intestine. Donkin et al. (2009) suggested that glycerol intake improves the efficiency of nitrogen utilization; however, this may be due to a reduction in urea blood levels, which was not observed in the present study (Table 5), but these authors did not observed influence of the glycerin intake over the microbial crude protein yield. Glycerol also positively influences milk CP concentration when cows face energy deficiency, independently of glycemia increase (Bodarski et al., 2005). On the other hand, those authors used glycerol as a supplement, whereas in the present study, CGL was added to the diet at the expense of corn, supplying equal energy levels (Table 1). Therefore, further studies are clearly needed to elucidate the mechanisms by which glycerol positively influences the efficiency of CP synthesis by the mammary gland.

4.3. Feed intake Donkin et al. (2009) observed feed intake reduction during the first seven days of glycerol feeding, possibly due to the adaptation of cows to the new feed. This was not observed in the present study, because of the 10 days adaptation period of the cows to the experimental diets. Therefore, it was shown that, after an adaptation period, CGL does not influence feed intake. Mach et al. (2009) also did not detect any differences in DMI when feeding 0, 40, 80, and 120 g CGL/kg of DM to Holstein bulls in feedlots.

4.4. Digestibility Cellulolytic microbial activity, and consequently, NDFD seems to suffer a strong negative influence of the dietary inclusion of high glycerol levels (Abo El-Nor et al., 2010; AbuGhazaleh et al., 2011; Donkin et al., 2009; Roger et al., 1992). However, when the obtained NDFD results are analyzed, it is important to take into account that CGL partially replaced corn that has high levels of starch, which has negative associative effects on fiber digestibility similar to glycerol, reducing NDFD (Chase & Hibberd, 1987), and in the present study, corn dietary levels were very high. For instance, in Control, it was 13.7 g/kg body weight. The results presented here demonstrate that these negative associative effects on fiber utilization are not potentiated by the CGL.


C.A. Wilbert et al. / Animal Feed Science and Technology 183 (2013) 116–123

4.5. Biochemical parameters In the present study, no influence of treatments was detected on NEFA serum concentrations, possibly because the experimental cows were in the middle third of lactation (65–173 days in milk, considering the entire experimental period), when feed intake traditionally reaches its peak (NRC, 2001). Therefore, cows with intermediate milk yield, as those used in the present study, would hardly suffer any severe energy deficit, and hence would not need to mobilize their body reserves. Urea serum concentration is considered a good indicator of dietary energy:RDP balance, and therefore, with diets containing equal protein and energy levels, there is no influence of CGL on that balance (Table 5). As previously mentioned, a large portion of the glycerol fermented in the rumen is transformed into propionic acid, which is the precursor of gluconeogenesis in ruminants. However, a large portion of CGL is absorbed intact, and participates in gluconeogenesis even more efficiently (Krehbiel, 2008). In the present study, glycemia presented a quadratic behavior in response to CGL intake (Fig. 1, P<0.05), with a reduction at the lower inclusion levels (G4 and G8) and subsequent increase (G12). The lower glycemia verified in G4 indicates that, in diets with low glycerin inclusion at the expense of corn, starch presents higher efficiency of conversion into glucose by ruminal fermentation and VFA generation possibly due to the characteristics of the microbial population in diets with high corn levels. Another hypothesis is that, there was higher glycerol fermentation in G4 and G8 than in G12, from which a larger portion escapes ruminal fermentation and it is absorbed intact. These hypothesis, however speculative, demonstrate the need for further in vivo studies on the fates of glycerol orally consumed, using C-labelled, for example. Rémond et al. (1993) using only four animals in in vitro and in vivo measurements estimated 0.25–0.55 of supplemental glycerol escapes ruminal fermentation in maize based silage diets, but was not observed influence of two levels of glycerol intake (240 or 1200 g/d). The authors conclusion is that with a higher quantity of glycerol, the proportion fermented relative to the proportion absorbed cannot be deduced with certainty from this trial. 5. Conclusions Crude glycerin is a good alternative energy feedstuff, and may be added up to 120 g/kg of total DM intake in partial replacement of ground corn grain in the diet of dairy cows, with no detrimental effects on milk yield, energy status, milk composition, feed intake or diet digestibility. Crude glycerin intake may increase milk protein content. Further in vivo studies are required to evaluate the full potential of crude glycerin as a feedstuff for lactating dairy cows, including the supply of inclusion levels higher than those evaluated in the present study. However, it is clear that crude glycerin is a very good alternative energy feedstuff for dairy cows. Acknowledgements The authors express their gratitude to Conselho Nacional de Pesquisa (CNPq) and to the company Granol S.A. for funding this study. References Abo El-Nor, S., AbuGhazaleh, A.A., Potu, R.B., Hastings, D., Khattab, M.S.A., 2010. Effects of differing levels of glycerol on rumen fermentation and bacteria. Anim. Feed Sci. Technol. 162, 99–105. AbuGhazaleh, A.A., Abo El-Nor, S., Ibrahim, S.A., 2011. The effect of replacing corn with glycerol on ruminal bacteria in continuous culture fermenters. J. Anim. Physiol. Anim. Nutr. 95, 313–319. Agência Nacional do Petróleo (ANP). Retrieved March 3, 2012, from: Association of Official Analytical Chemists (AOAC), 1996. Official Methods of Analysis, 16th ed. AOAC, Washington, DC, USA. Bodarski, R., Wertelecki, T., Bommer, F., Gosiewski, S., 2005. The changes of metabolic status and lactation performance in dairy cows under feeding tmr with glycerin (glycerol) supplement at periparturient period. Electron. J. Pol. Agric. Univ., Anim. Husb. 8, 1–9. Carvalho, E.R., Schmelz-Roberts, N.S., White, H.M., Doane, P.H., Donkin, S.S., 2011. Replacing corn with glycerol in diets for transition dairy cows. J. Dairy Sci. 94, 908–916. Celik, E., Ozbay, N., Oktar, N., Calk, P., 2008. Use of biodiesel byproduct crude glycerol as the carbon source for fermentation processes by recombinant Pichia pastoris. Ind. Eng. Chem. Res. 47, 2985–2990. Chase, C.C., Hibberd, C.A., 1987. Utilization of low-quality native grass hay by beef cows fed increasing quantities of corn grain. J. Anim. Sci. 65, 557–566. Chung, Y.H., Rico, D.E., Martinez, C.M., Cassidy, T.W., Noirot, V., Ames, A., Varga, G.A., 2007. Effects of feeding dry glycerin to early postpartum Holstein dairy cows on lactational performance and metabolic profiles. J. Dairy Sci. 90, 5682–5691. Dasari, M.A., Kiatsimkul, P.P., Sutterlin, W.R., Suppes, G.J., 2005. Low-pressure hydrogenolysis of glycerol to propylene glycol. Appl. Catal. A: Gen. 281, 225–231. DeFrain, J.M., Hippen, A.R., Kalscheur, K.F., Jardon, P.W., 2004. Feeding glycerol to transition dairy cows: effects on blood metabolites and lactation performance. J. Dairy Sci. 87, 4195–4206. Donkin, S.S., Doane, P., 2007. Glycerol as a feed ingredient in dairy rations. In: Proceeding from the 2007 Tri-State Dairy Nutrition Conference, Fort Wayne, IN. The Ohio State University, Columbus, pp. 97–103. Donkin, S.S., Koser, S., White, H., Doane, P.H., Cecava, M.J., 2009. Feeding value of glycerol as a replacement for corn grain in rations fed to lactating dairy cows. J. Dairy Sci. 92, 5111–5119. Easley, J.F., McCall, J.T., Davies, G.K., Shirley, R.L., 1965. Analytical Methods for Feeds and Tissues. Nutrition Laboratory, Department of Animal Science, University of Florida, Gainesville, pp. 81. Krehbiel, C.R., 2008. Ruminal and physiological metabolism of glycerin. J. Anim. Sci. (E-Suppl. 2) 86, 392. Lage, J.F., Paulino, P.V., Pereira, L.G.R., Valadares, F S.C., de Oliveira, A.S., Detmann, E., Souza, N.K.P., Lima, J.C.M., 2010. Glicerina bruta na dieta de cordeiros terminados em confinamento. Pesq. Agrop. Bras. 45, 1012–1020 (Eng. Abstr.). Mach, N., Bach, A., Devant, M., 2009. Effects of crude glycerin supplementation on performance and meat quality of Holstein bulls fed high-concentrate diets. J. Anim. Sci. 87, 632–638, in press.

C.A. Wilbert et al. / Animal Feed Science and Technology 183 (2013) 116–123


Mackle, T.R., Dwyer, D.A., Ingvartsen, K.L., Chouinard, P.Y., Ross, D.A., Bauman, D.E., 2000. Effects of insulin and postruminal supply of protein on use of amino acids by the mammary gland for milk protein synthesis. J. Dairy Sci. 83, 93–105, in press. National Research Council (NRC), 2001. Nutrient Requirements of Dairy Cattle, 7th rev. ed. National Academic Press, Washington, DC. Organisation for Economic Co-operation and Development – Food and Agricultural Organization (OECD-FAO), 2010. OECD-FAO agricultural outlook 2010–2019, OECD, Food and Agriculture Organization of the United Nations. OECD Publishing, Paris, FR. Pachauri, N., He, B., 2006. Value-added utilization of crude glycerol from biodiesel production: a survey of current research activities. In: Proceeding from the 2006 ASABE Annual International Meeting, Portland, OR. The American Society of Agricultural and Biological Engineers, St. Joseph, MC, pp. 1–16. Rémond, B., Souday, E., Jouany, J.P., 1993. In vitro and in vivo fermentation of glycerol by rumen microbes. Anim. Feed Sci. Technol. 41, 121–132. Roger, V., Fonty, G., Andre, C., Gouet, P., 1992. Effects of glycerol on the growth, adhesion and cellulolytic activity of rumen cellulolytic bacteria and anaerobic fungi. Curr. Microbiol. 25, 197–201. SAS, 2002. Statistical Analysis Systems Institute, Version 9. 0. SAS Institute Inc, Cary, NC (USA). Schröder, A., Südekum, K.H., 1999. Glycerol as a by-product of biodiesel production in diets for ruminants. In: Proceedings of the 10th International Rapeseed Congress, Canberra, Australia, September, 26–29, Paper No. 241. Sjaunja, L.O., Baevre, L., Junkkarinen, L., Pedersen, J., Setälä, J., 1990. A Nordic proposal for an energy corrected milk (ECM) formula. In: Proceedings of the 27th Session of International Committee of Recording and Productivity of Milk Animal, July 2–6, Paris, France, pp. 156–157. Van Soest, P.J., Robertson, J.B., Lewis, B.A., 1991. Methods for dietary fiber, neutral detergent fiber, and nonstarch polysaccharides in relation to animal nutrition. J. Dairy Sci. 74, 3583–3597. Van Soest, P.J., 1994. Nutritional Ecology of the Ruminant, 2nd ed. Cornell University Press, Ithaca, pp. 476. Wang, C., Liu, Q., Huo, W.J., Yang, W.Z., Dong, K.H., Huang, Y.X., Guo, G., 2009. Effects of glycerol on rumen fermentation, urinary excretion of purine derivatives and feed digestibility in steers. Livest. Sci. 121, 15–20. Zeoula, L.M., Martins, A.S., Prado, I.N., Alcalde, C.R., Branco, A.F., Santos, G.T., 1999. Solubilidade e degradabilidade ruminal do amido de diferentes alimentos. Rev. Bras. Zootec. 28, 898–905 (Eng. Abstr.).