Polyurethane Foams for Thermal Insulation Uses Produced from Castor Oil and Crude Glycerol Biopolyols

Rigid polyurethane foams were synthesized using a renewable polyol from the simple physical mixture of castor oil and crude glycerol. The effect of the catalyst (DBTDL) content and blowing agents in the foams’ properties were evaluated. The use of physical blowing agents (cyclopentane and n-pentane) allowed foams with smaller cells to be obtained in comparison with the foams produced with a chemical blowing agent (water). The increase of the water content caused a decrease in density, thermal conductivity, compressive strength, and Young’s modulus, which indicates that the increment of CO2 production contributes to the formation of larger cells. Higher amounts of catalyst in the foam formulations caused a slight density decrease and a small increase of thermal conductivity, compressive strength, and Young’s modulus values. These green foams presented properties that indicate a great potential to be used as thermal insulation: density (23–41 kg·m−3), thermal conductivity (0.0128–0.0207 W·m−1·K−1), compressive strength (45–188 kPa), and Young’s modulus (3–28 kPa). These biofoams are also environmentally friendly polymers and can aggregate revenue to the biodiesel industry, contributing to a reduction in fuel prices.


Introduction
Rigid polyurethane foams are usually applied as thermal insulation in buildings, as well as in automobile and aerospace industries. The global rigid polyurethane foam market is expected to reach USD $20.40 billion by 2020 in construction applications, such as in residential and commercial roofs, walls, panels, and doors, and in appliance applications [1]. For insulation purposes, thermal conductivity is the most important property to be studied. Typical thermal conductivity values for polyurethane foams are between 0.02 and 0.03 W·m −1 ·K −1 and the very efficient thermal insulation of these materials is due to the extremely low thermal conductivity of the blowing agent gas trapped in the closed porous structures [2,3]. Additives, as blowing agents, catalysts, and surfactants, are very important to adjust the final properties of the foam's synthesis from the main reaction of an isocyanate with a polyol. However, most of these reagents are derived from petrochemicals, increasing the petroleum dependence and, as a consequence, environmental problems. These aspects have encouraged the production of rigid foams from renewable materials [4].

Study of the Best Binary Mixture to Produce Polyurethane Foams
A preliminary study to choose the best polyol formulation was carried out. For this purpose, rigid polyurethane foams were synthesized using the batch process method and binary polyols with different compositions. This comparative study was performed with the mixture of P.A. glycerol (20%, 40%, 50%, 60%, 80% w/w) and castor oil, and with crude glycerol (10%, 20%, 30%, 40%, 50%, 60%, 70%) and also castor oil. The NCO/OH molar ratio was kept equal to 2.0, the surfactant, 2% (w/w), and water was employed as the blowing agent (2%). These mixtures were kept under vigorous stirring for 1 min at 500 rpm. Afterwards, the isocyanate was added to the pre-mixture and stirred until complete homogenization. The formulations were poured into a wooden mold with the dimensions 7.0 × 7.0 × 20.0 cm for the growth of the polymer foams and cured for 24 h at room temperature. After the curing time, the foams were demolded and visual inspections were performed. The best polyol formulation was chosen for the foam with the best properties (aspect, homogeneity, dimensional stability, and low friability).

Study of Catalyst and Blowing Agent Effect on the Foams Properties
After preliminary tests, the best binary polyol GCo with 10% crude glycerol, and 90% castor oil w/w was used to produce foams with good thermal stability, mechanical properties, and low thermal conductivity for insulation use. For this purpose, this best polyol was tested to prepare foams aiming to evaluate the effect of different catalyst (DBTDL) amounts (1% and 2% w/w); different blowing agent types (water, n-pentane, and cyclopentane), and contents. The tests to evaluate the different blowing agent amounts were carried out with: water = 2%, 4% and 6% w/w; n-pentane = 2% w/w; and cyclopentane = 2%, 4% and 6% w/w, but 6% was not suitable for use. These foam reactions were performed using the same methodology described for the preliminary tests (Section 2.3.1). These formulations are summarized in Table 1.

FTIR Analyses
Fourier transform infrared (FTIR) spectra were recorded on an ABB Bomer spectrometer using 0.1% KBr pellets and an ATR accessory with resolution of 4 cm −1 , in the range of 4000-400 cm −1 .

Thermogravimetric Analyses (TGA and DTG)
The thermal behavior of the foams was evaluated by thermogravimetric analysis (TGA), which was carried out on a TA equipment model TA-50 Q at temperatures ranging from 30 • C to 800 • C with a heating rate of 10 • C·min −1 under a nitrogen flow of 40 mL·min −1 .

Morphology Analyses
The images obtained by optical microscopy were collected on an Olympus optical microscope, model BX41M, coupled to a TecVoz camera, model DNS 480. The morphology of polymers structure was characterized to observe the foam cellular structure by scanning electron microscope (SEM) images, which were obtained with a JSM-6360LV, a JEOL scanning microscope.

Determination of Apparent Density
Polyurethane foam densities were calculated according to the ASTM D-1622 standard by measuring three specimens of each sample with the dimensions of 50 × 50 × 25 mm.

Mechanical Properties
Mechanical testing (10% compressive strength) of polyurethane foams was performed on an Autograph Precision Universal Testing Machine AG-Xplus Series (Shimadzu, Kyoto, Japan), according to the ASTM D1621 standard method. At least five samples were analyzed to obtain the average value; the specimen size was 50 × 50 × 25 mm.

Thermal Conductivity
Thermal conductivity tests were performed by an instrument developed by Professor Vagner Carvalho in the Surface Laboratory at the Physics Department at Universidade Federal de Minas Gerais-Brazil [29]. This equipment has a unit that contains a tip, which is electronically controlled and has a temperature sensor. The measurements are based on the thermal comparison method by copper-constantan differential thermocouples that measure the contact temperature gradient and the changes in the sample temperature. The samples were tested six times and a simple average was taken as the comparator output. A calibration curve was constructed using various ranges of samples with a wide thermal conductivity and the data were plotted in order to obtain the relation between the thermal conductivity and the comparator output.

Study of the Best Binary Mixture to Produce Polyurethane Foams
The binary polyol production was first studied by the physical mixture of pure glycerol and castor oil, varying the glycerol content. Some foams did not present good dimensional stability (Figure 1a). Upon increasing the content of pure glycerol, it was observed that the foams became denser and softer. The formulations with 20% and 40% (w/w) of pure glycerol content did not grow as a typical foam, yielding a very rigid, solid material. The foams produced with the polyol containing 50% (w/w) of pure glycerol presented a high homogeneity, but upon increasing this content, the foams became very friable. crude glycerol for producing polyols from biomass liquefaction. These studies also affirm that these impurities of crude glycerol improved the polyols' and polyurethanes' properties [9,11,30]. It was observed that upon increasing the amount of crude glycerol, there was a decrease in the rigidity and the dimensional stability of the foams. For this reason, the foam produced from the polyol containing 10% of crude glycerol and 90%of castor oil (w/w) (Figure 1f) was chosen to perform further studies. The hydroxyl number (240 mg•KOH•g −1 ) and viscosity (436.5 mm 2 •s −1 ) of this polyol were measured, indicating that these polyols are adequate to produce rigid foams [4]. Similar results have already reported in the literature for polyols from castor oil [26].
It is important to point out that the polyol used to produce our best foam, with 10% of glycerol and 90% of castor oil (w/w), has the molar ratio glycerol/castor oil approximately equal to 1 (considering the molar weight of glycerol and castor oil 92.09 and 895.33 g•mol −1 , respectively). Observing the structure of these molecules (Figure 2), there are three hydroxyl groups in each glycerol molecule and three instaurations of ricinoleic acid in the triglyceride structure that are suitable to be transformed into OH groups by pre-treatment reactions. Thus, we can consider that 1 mol of glycerol has the same number of OH groups as 1 mol of castor oil. Then, when we used a binary mixture 1:1, the OH amount is doubled. The same value of hydroxyl groups can be obtained by inserting an OH on each instauration of the castor oil ricinoleic chain. Our study was then performed using the binary mixture, without castor oil modification, to avoid extra costs in the process.  The pure glycerol was then replaced by crude glycerol, a co-product of biodiesel production, in order to synthesize new foams and the results were quite different. The foams with crude glycerol and castor oil polyol (named GCo, Figure 1f-l) were more homogeneous and presented good dimensional stability in comparison to those synthesized with pure glycerol (Figure 1a-e). Based on these experimental behaviors, we believe that the crude glycerol impurities (alkaline catalyst, methanol, methyl esters of fatty acids, fatty acid methyl esters) are responsible for the best properties of the foams. Further investigations can be made to understand this behavior. Similar behaviors have already been reported in the literature, evaluating the effects of the replacement of pure glycerol by crude glycerol for producing polyols from biomass liquefaction. These studies also affirm that these impurities of crude glycerol improved the polyols' and polyurethanes' properties [9,11,30].
It was observed that upon increasing the amount of crude glycerol, there was a decrease in the rigidity and the dimensional stability of the foams. For this reason, the foam produced from the polyol containing 10% of crude glycerol and 90%of castor oil (w/w) (Figure 1f) was chosen to perform further studies. The hydroxyl number (240 mg·KOH·g −1 ) and viscosity (436.5 mm 2 ·s −1 ) of this polyol were measured, indicating that these polyols are adequate to produce rigid foams [4]. Similar results have already reported in the literature for polyols from castor oil [26].
It is important to point out that the polyol used to produce our best foam, with 10% of glycerol and 90% of castor oil (w/w), has the molar ratio glycerol/castor oil approximately equal to 1 (considering the molar weight of glycerol and castor oil 92.09 and 895.33 g·mol −1 , respectively). Observing the structure of these molecules ( Figure 2), there are three hydroxyl groups in each glycerol molecule and three instaurations of ricinoleic acid in the triglyceride structure that are suitable to be transformed into OH groups by pre-treatment reactions. Thus, we can consider that 1 mol of glycerol has the same number of OH groups as 1 mol of castor oil. Then, when we used a binary mixture 1:1, the OH amount is doubled. The same value of hydroxyl groups can be obtained by inserting an OH on each instauration of the castor oil ricinoleic chain. Our study was then performed using the binary mixture, without castor oil modification, to avoid extra costs in the process. glycerol molecule and three instaurations of ricinoleic acid in the triglyceride structure that are suitable to be transformed into OH groups by pre-treatment reactions. Thus, we can consider that 1 mol of glycerol has the same number of OH groups as 1 mol of castor oil. Then, when we used a binary mixture 1:1, the OH amount is doubled. The same value of hydroxyl groups can be obtained by inserting an OH on each instauration of the castor oil ricinoleic chain. Our study was then performed using the binary mixture, without castor oil modification, to avoid extra costs in the process.

Study of Catalyst and Blowing Agent Effect in the Foams Properties
The characterization data obtained for the different foams, which were prepared using the best binary polyol (10% crude glycerol and 90% castor oil w/w) will be discussed in this section. The formulations will be presented using Roman numerals, as shown in Table 1.
The FTIR spectra of the renewable raw materials used to produce GCo polyols are shown in Figure 3. The band corresponding to the hydroxyl group's vibration is observed at approximately 3700-3000 cm −1 . The characteristic stretches of double bonds in the castor oil groups C=C-H and C=C are observed at 3020 and 1740 cm −1 , respectively. The bands around 3018 and 2710 cm −1 are assigned to CH 2 and CH 3 stretches of aliphatic chains, which are quite pronounced in castor oil due to the 18-carbon chain. The characteristic band of carbonyl and carboxyl groups is observed to be centered at 1743 cm −1 in the castor oil spectrum. The alkenes deformation of CH 2 groups, present in the castor oil structure, is observed in a strong band at 1458 cm −1 . The bands around 1112-1000 cm −1 indicate the presence of primary and secondary hydroxyl groups. These bands are very pronounced in crude glycerol spectrum, due to the three hydroxyl groups present in its short chain [16,18].

Study of Catalyst and Blowing Agent Effect in the Foams Properties
The characterization data obtained for the different foams, which were prepared using the best binary polyol (10% crude glycerol and 90% castor oil w/w) will be discussed in this section. The formulations will be presented using Roman numerals, as shown in Table 1.
The FTIR spectra of the renewable raw materials used to produce GCo polyols are shown in Figure 3. The band corresponding to the hydroxyl group's vibration is observed at approximately 3700-3000 cm −1 . The characteristic stretches of double bonds in the castor oil groups C=C-H and C=C are observed at 3020 and 1740 cm −1 , respectively. The bands around 3018 and 2710 cm −1 are assigned to CH2 and CH3 stretches of aliphatic chains, which are quite pronounced in castor oil due to the 18-carbon chain. The characteristic band of carbonyl and carboxyl groups is observed to be centered at 1743 cm −1 in the castor oil spectrum. The alkenes deformation of CH2 groups, present in the castor oil structure, is observed in a strong band at 1458 cm −1 . The bands around 1112-1000 cm −1 indicate the presence of primary and secondary hydroxyl groups. These bands are very pronounced in crude glycerol spectrum, due to the three hydroxyl groups present in its short chain [16,18].   Table 1) All of the foams' spectra produced from GCo polyols are very similar, while a typical foam spectrum is shown in Figure 3, which presents the characteristic polyurethane bands. The stretching and vibrations of NH groups were observed between 3808-3308, and 1512 and 1510 cm −1 , respectively. The deformation of CH2 bonds was observed by the two thin bands at 2900 and 2890 cm −1 . The vibration of N=C=N and N=C=O groups are attributed to bands between 2390 and 2150 cm −1 . Other vibration modes of the CH bond were also observed at 1464, 1418, 1364, and 1294 cm −1 . The band between 1730 and 1720 cm −1 corresponds to the stretch of the CO-free urethane bond, and around 1700 cm −1 , the hydrogen bond between the carbonyl and hydrogen atoms (from NH groups) from urethane is also observed. A band related to stretching asymmetric links OCONH was revealed  Table 1) All of the foams' spectra produced from GCo polyols are very similar, while a typical foam spectrum is shown in Figure 3, which presents the characteristic polyurethane bands. The stretching and vibrations of NH groups were observed between 3808-3308, and 1512 and 1510 cm −1 , respectively. The deformation of CH 2 bonds was observed by the two thin bands at 2900 and 2890 cm −1 . The vibration of N=C=N and N=C=O groups are attributed to bands between 2390 and 2150 cm −1 . Other vibration modes of the CH bond were also observed at 1464, 1418, 1364, and 1294 cm −1 . The band between 1730 and 1720 cm −1 corresponds to the stretch of the CO-free urethane bond, and around 1700 cm −1 , the hydrogen bond between the carbonyl and hydrogen atoms (from NH groups) from urethane is also observed. A band related to stretching asymmetric links OCONH was revealed at 1380 cm −1 . The bands between 1100 and 1000 cm −1 were attributed to primary and secondary hydroxyl groups [16,17].
The thermal behavior of the GCo foams containing different types and amounts of catalyst, shown in Figure 4, were evaluated by thermogravimetric analysis (TGA and DTG). The different foams presented similar thermal stability, and the DTG curves showed three regions of weight loss. The first event (around 300 • C) corresponds to urethane thermal degradation, free isocyanate, and alcohols; the second event is related to the degradation of rigid segments, at 370 • C; and the third event, approximately at 480 • C, is associated to the thermal degradation of the flexible segments and others segments of the remaining structure [31,32].   Table 1.
The effect of different blowing agents on the thermal stability of the GCo foams was evaluated, as shown in Figure 4a,b. The results indicate that the blowing agent type did not significantly modify the thermal behavior of the foams, which is suggested by the similar curves of the foams synthesized with water, cyclopentane, and n-pentane.
The effect of the amount of blowing agent (water) in the formulations (Figure 4c,d) was also investigated. The results show that the amount of water as a blowing agent did not significantly affect the thermal stability of the foams produced with polyol GCo, taking into account that all of the  Table 1. The effect of different blowing agents on the thermal stability of the GCo foams was evaluated, as shown in Figure 4a,b. The results indicate that the blowing agent type did not significantly modify the thermal behavior of the foams, which is suggested by the similar curves of the foams synthesized with water, cyclopentane, and n-pentane.
The effect of the amount of blowing agent (water) in the formulations (Figure 4c,d) was also investigated. The results show that the amount of water as a blowing agent did not significantly affect the thermal stability of the foams produced with polyol GCo, taking into account that all of the curves presented the same profile, indicating a similar thermal stability.
Apparent density is an important parameter of cellular polymers. The effect of the blowing agent type on the apparent density of the foams produced from GCo polyols ( Table 2) showed that the formulations with physical blowing agents (cyclopentane and n-pentane) produced foams with higher densities than those synthesized with the chemical blowing agent (water). Similar results have been reported in the literature [32][33][34], and this behaviour indicates that smaller cells are formed due to the rapid volatilization of physical blowing agents, which have a low boiling point, during the highly exothermic foam growth step in comparison with the CO 2 , produced by the reaction of water with isocyanate [35]. The effect of the blowing agent (water) content on the foams' apparent densities was also evaluated, as shown in Figure 5a. In increasing the water amount, a decrease of density is observed, which suggests that higher cells are formed with the enhancement of CO 2 production from the water and isocyanate reaction [36].  The effect of the blowing agent (water) content on the foams' apparent densities was also evaluated, as shown in Figure 5a. In increasing the water amount, a decrease of density is observed, which suggests that higher cells are formed with the enhancement of CO2 production from the water and isocyanate reaction [36].  (Table 1) are indicated in each point of these graphics. Figure 5a also shows the influence of the catalyst content on the foams density. A decrease of apparent density was observed upon increasing the catalyst amount in the formulations. This behavior can be explained by the increment in the polymerization rate with the enhancement of the organometallic catalyst content in the formulation, avoiding the CO2 releasing during the foam's cell formation [4]. As the reaction occurs with higher speed, the blowing agent is trapped in the structure and the cells presented higher diameters and lower densities (Figure 5a,b, respectively) [37]. This effect is more prominent in foams with higher water contents. These apparent density results are in agreement of the values measured for the same rigid polyurethane foams synthesized using castor oil polyols [19,26].
The effect of different blowing agents on the cellular foam structures can also be observed in  formation [4]. As the reaction occurs with higher speed, the blowing agent is trapped in the structure and the cells presented higher diameters and lower densities (Figure 5a,b, respectively) [37]. This effect is more prominent in foams with higher water contents. These apparent density results are in agreement of the values measured for the same rigid polyurethane foams synthesized using castor oil polyols [19,26].
The effect of different blowing agents on the cellular foam structures can also be observed in Figure 6, which shows SEM images of the foams synthesized with water and cyclopentane. The foams prepared with water as the blowing agent showed the largest cell size, confirming the density data (Figure 5a). Pentane has a low boiling temperature (around 50 • C) and volatizes very quickly, as previously explained in the density data discussion. The foam with 6% cyclopentane presented a low dimensional stability and, for this reason, its SEM micrograph was not shown here. water. The foams formulated with 6% water produced larger and heterogeneous cells, indicating that 4% water is the optimal quantity to be used in the foam formulations.   (Table 1) are indicated in each micrograph.
The comparison of the catalyst amount in the foam cells ( Figure 6) formulated with water showed that an increment of the catalyst content yielded cellular materials with higher average cell diameters, confirming the density values in Figure 5a. The foams synthesized with 2% DBTDL showed the best cell homogeneity, despite higher cell diameters, as observed in Figure 5b. The average diameter of the foams produced in this present process is smaller than the data reported in The foams formulated with water as the blowing agent presented the best dimensional stability, the lowest apparent density, and the higher cells homogeneity. Based on these results, we chose this formulation, aiming to evaluate the effect of the catalyst amount on the mechanical and conductivity properties. Another important aspect to note is that the use of water as a blowing agent is considered a green and inexpensive option.
The effect of the water content as the blowing agent was also evaluated by SEM images, as shown in Figure 6. It was observed that the water concentration is directly proportional to the cell size (Figure 5b). These analyses are in agreement with the density data (Figure 5a). The foams produced with 4% water presented higher cell homogeneity in comparison to those containing 2% water. The foams formulated with 6% water produced larger and heterogeneous cells, indicating that 4% water is the optimal quantity to be used in the foam formulations.
The comparison of the catalyst amount in the foam cells ( Figure 6) formulated with water showed that an increment of the catalyst content yielded cellular materials with higher average cell diameters, confirming the density values in Figure 5a. The foams synthesized with 2% DBTDL showed the best cell homogeneity, despite higher cell diameters, as observed in Figure 5b. The average diameter of the foams produced in this present process is smaller than the data reported in the literature (from 107 to 121 µm) of foams synthesized from pre-polymerized castor oil [28], which is an important result for our foam uses.
The main property for the application of foam as thermal insulation is its thermal conductivity. This parameter was measured for the rigid foams synthesized with water as a blowing agent, and the results are presented in Figure 7. It was observed that upon increasing the water amount in these formulations, there was a decrease in the thermal conductivity. This result can be explained by the decreasing density and the enhancement of the average cell diameter of the foams [38].  The effect of the catalyst amount on this property is also presented in Figure 7. The use of a higher catalyst content in the formulations causes a slight thermal conductivity value enhancement, despite the density decrease as a consequence of cell size increase, as shown in Figure 5. The foams synthesized in this study presented better results in comparison to those reported in the literature for foams derived from renewable raw materials, whose values vary between 0.0233 and 0.0505 W•m −1 •K −1 , suggesting that these materials have a potential use as thermal insulation [22,39,40]. These thermal conductivity results are also better than those found for foams produced from pre-treated castor oil, especially if we consider the use of the very simple and inexpensive method of production [19,28].
The mechanical properties of the foams synthesized with different contents of blowing agent and catalyst were evaluated, and the results are shown in Figure 8. These results presented similar values to those reported in the literature of foams produced from castor oil polyols, which ranged from 125 to 220 kPa [16,19,25,26]. A significant decrease in the compressive strength and Young's modulus of the foams were observed with the addition of higher blowing agent amounts, which can The effect of the catalyst amount on this property is also presented in Figure 7. The use of a higher catalyst content in the formulations causes a slight thermal conductivity value enhancement, despite the density decrease as a consequence of cell size increase, as shown in Figure 5. The foams synthesized in this study presented better results in comparison to those reported in the literature for foams derived from renewable raw materials, whose values vary between 0.0233 and 0.0505 W·m −1 ·K −1 , suggesting that these materials have a potential use as thermal insulation [22,39,40]. These thermal conductivity results are also better than those found for foams produced from pre-treated castor oil, especially if we consider the use of the very simple and inexpensive method of production [19,28].
The mechanical properties of the foams synthesized with different contents of blowing agent and catalyst were evaluated, and the results are shown in Figure 8. These results presented similar values to those reported in the literature of foams produced from castor oil polyols, which ranged from 125 to 220 kPa [16,19,25,26]. A significant decrease in the compressive strength and Young's modulus of the foams were observed with the addition of higher blowing agent amounts, which can be related with the decrease in density and the increase in cell size. As the cell structure becomes higher, less force is necessary to cause deformation in these foams [36]. despite the density decrease as a consequence of cell size increase, as shown in Figure 5. The foams synthesized in this study presented better results in comparison to those reported in the literature for foams derived from renewable raw materials, whose values vary between 0.0233 and 0.0505 W•m −1 •K −1 , suggesting that these materials have a potential use as thermal insulation [22,39,40]. These thermal conductivity results are also better than those found for foams produced from pre-treated castor oil, especially if we consider the use of the very simple and inexpensive method of production [19,28].
The mechanical properties of the foams synthesized with different contents of blowing agent and catalyst were evaluated, and the results are shown in Figure 8. These results presented similar values to those reported in the literature of foams produced from castor oil polyols, which ranged from 125 to 220 kPa [16,19,25,26]. A significant decrease in the compressive strength and Young's modulus of the foams were observed with the addition of higher blowing agent amounts, which can be related with the decrease in density and the increase in cell size. As the cell structure becomes higher, less force is necessary to cause deformation in these foams [36]. The results of the compressive strength and Young's modulus of foams with different amounts of catalyst in the formulations (Figure 8a,b) showed that there are no significant changes in the values upon increasing the catalyst amount, especially for the formulations with 4% and 6% of water as the blowing agent. The variations are within the experimental errors.
Comparing all of the formulations, it was observed that the foam with best thermal conductivity (0.0141 W·m −1 ·K −1 ) was formulated with 1% of DBTDL and 6% of water, which also showed a low apparent density value (23.9 kg·m −3 ). However, this sample showed a low compressive strength (51.01 kPa) and Young's modulus (3.44 kPa), suggesting its application as an insulator of places that do not receive high loads. The foam containing 2% of DBTDL and 2% of water possess a higher compressive strength (187.93 kPa) and Young's modulus (27.74 kPa), as well as a low apparent density value (37.4 kg·m −3 ). On the other hand, the thermal conductivity value was the higher (0.0207 W·m −1 ·K −1 ) in comparison with the other formulations; indeed, this insulation property value is in the range of typical commercial products [2].

Conclusions
Rigid polyurethane foams were synthesized from a polyol produced by the physical mixture of castor oil and crude glycerol, a co-product from the biodiesel industry. The polyol preparation method is simple and does not require any pre-treatment of the raw materials. The best formulation was obtained with 10% of crude glycerol and 90% of castor oil (w/w, molar ratio = 1:1) and water as the blowing agent. The addition of 10% glycerol increased the OH content of the biopolyol to the same extent as if the castor oil had been pre-treated in order to insert one hydroxyl group in each instauration (C=C) of ricinoleic acid.
The effect of the blowing agent on the foam synthesis was evaluated and the results showed that increasing the water amount caused a decrease in density, thermal conductivity, compressive strength, and Young's modulus. These behaviors are due to the cell size enhancement.
The evaluation of the effect of the catalyst (DBTDL) amount showed that increasing the catalyst content caused a decrease in density and an increase in thermal conductivity, however, there was only a slight influence on the compressive strength and Young's modulus.
These innovative foams presented properties that indicate their great potential to be used as thermal insulation. These materials are inexpensive, environmentally friendly, and can contribute to reducing the biodiesel prices in a biorefinery approach. These biopolymers can be synthesized by a simple process, which can be easily used in a large scale.