Materials and Design 110 (2016) 526–531

Contents lists available at ScienceDirect

Materials and Design

j ourna l homepage: www.e lsev ie r .com/ locate /matdes

Dispersion and reinforcing effect of carrot nanofibers onbiopolyurethane foams

Xiaojian Zhou a, Jatin Sethi a, Shiyu Geng a, Linn Berglund a, Nikolina Frisk a, Yvonne Aitomäki a,Mohini M. Sain a,b, Kristiina Oksman a,b,⁎a Division of Materials Science, Luleå University of Technology, 971 87 Luleå, Swedenb Centre of Biocomposite and Biomaterials Processing, Faculty of Forestry, University of Toronto, 33 Willcocks Street, Toronto M5S 3B3, Canada

H I G H L I G H T S G R A P H I C A L A B S T R A C T

• Semirigid castor oil based PU foams withvery low bulk density (b50 kg/m3) weresuccessfully prepared.

• Carrot nanofibers were used as rein-forcement to improve cell wall rigidity.

• Prepared nanocomposite foams areperforming in the level of commercialPU foams.

• Foams were shown to be excellent corefor biocomposite laminates.

⁎ Corresponding author at: Division of MaterialsTechnology, 971 87 Luleå, Sweden.

E-mail address: kristiina.oksman@ltu.se (K. Oksman).

http://dx.doi.org/10.1016/j.matdes.2016.08.0330264-1275/© 2016 Elsevier Ltd. All rights reserved.

a b s t r a c t

a r t i c l e i n f o

Article history:Received 8 July 2016Received in revised form 1 August 2016Accepted 9 August 2016Available online 10 August 2016

In this study, carrot nanofibers (CNF) were used to enhance the performance of biobased castor oil polyol poly-urethane nanocomposite foams. A method of dispersing CNF in the polyol was developed and the foam charac-teristics and CNF reinforcing effect were studied. Co-solvent-assisted mixing resulted in well-dispersed CNF inthe polyol, and foams with 0.25, 0.5 and 1 phr CNF content were prepared. The reinforced nanocompositefoams displayed a narrow cell size distribution and the compressive strength andmodulus were significantly el-evated and the best compressive strength andmoduluswere reachedwith 0.5 phr CNF. Similarly, themodulus ofthe solid material was also significantly increased based on theoretical calculations. When comparing the foamperformance, compressive strength and stiffness as a function of the density, the nanocomposite foams performsas commercial rigid PU foam with a closed cell structure. These results are very promising and we believe thatthese foams are excellent core materials for lightweight sandwich composites.

© 2016 Elsevier Ltd. All rights reserved.

Keywords:Polyurethane foamCastor oil polyolCellulose nanofibersDispersionCompressive mechanical properties

1. Introduction

During the past 10–15 years, the utilization of bioresources to pre-pare polyurethane (PU) foams has attracted attention because of the

Science, Luleå University of

increasing concerns about environmental problems, replacement of fos-sil resources and the lowprice and abundant supply of natural vegetableoils and biomass by-products [1–7]. However, the prepared biobasedpolyurethane (BPU) foams are not yet suitable for commercial applica-tions because they do notmeet the industrial standard requirements forcompression strength (≥180 kPa) [1,4]. Therefore, enhancing the prop-erties of BPU foamhas been and continues to be of interest. Many differ-ent strategies have been used to improve the properties of the foams.

527X. Zhou et al. / Materials and Design 110 (2016) 526–531

One strategy is to use different vegetable oils with different reactivity,where the polyol is modified to a higher hydroxyl value factor [3,4]. An-other is the optimization of the isocyanate index [5]. However, theresulting increase in mechanical strength is limited by the optimizationof the formulation and mass proportion. Different reinforcements, toimprove the mechanical properties of the foams, have also been tested[8–15]. Gu et al. [9] tested hardwood pulp and Silva et al. [10] celluloseresidue from pulp industry but the use of these fibers did not result inimproved mechanical properties. Mosiewicki et al. [12] added woodflour, 5, 10 and 15 wt% into the castor oil based PU, but the mechanicalproperties of the foamdecreasedwith the increasedwood flour content.Zhu et al. [8] used cellulose fiberswith a smaller size and reported an in-crease of compressive strength of soy-oil based foam from approx.110 kPa to 170 kPa. Luo et al. [15] used lignin as reactive reinforcementin BPU foams and showed that the addition of 10 wt% lignin improvedthe foam structure as well as its mechanical properties.

Nano-sized reinforcements have also been of interest, since theirsmall size should allow reinforcement of the cell walls and struts ofthe foam resulting in better compressive and flexural properties. Zhuet al. [8] reported that the use of surface modified montmorillonitenanoclay into a soy-oil BPU foam improved its mechanical propertieshowever it increased the density of the foam. Biobasednanoreinforcements such as cellulose nanofibers [16] and nanocrystals[17–21] have also been tested. Typically the cellulose nanomaterialsused have been freeze-dried and then mixed with the polyol. However,it is difficult to disperse the dried nanocellulose into the hydrophobicpolyol. Several procedures have been tested to improve the dispersion.Li et al. [17] dispersed freeze-dried cellulose nanowhiskers (crystals)first in dimethylformamide (DMF) solvent by sonication and then com-bined this mixture with polyol and the solvent was removed using vac-uum. Improved compression strength and stiffness as well as a morehomogenous cell size and improved thermal stability were reported.To avoid freeze drying of the nanocellulose and the use of the solvent,the same authors [18] directlymixed aqueous nanocellulose suspensionand polyol, followed by removing the water under vacuum, and thenmixed in other foaming components. This procedure resulted in betterdispersion and further improved properties. Cordero et al. [21] used acombination of sonication and homogenization processes to dispersethe cellulose nanocrystals (CNC) in a polyol but unfortunately they didnot report any compressive strength ormodulus values for the preparedfoams. However in the attempts to disperse the nanocelluloses inpolyol, detailed information of the dispersion of nanocellulose in polyolhas not been shown in these studies [16–21].

We showed in an earlier study [22] that carrot nanofibers (CNF),separated from residue from raw juice production, have similar andeven better mechanical properties to many other cellulose nanofibers.These carrot nanofibers have been shown to be more easily re-dispers-ible, are low in cost and can be separated more energy efficiently thanmany other raw materials. These reasons make carrot nanofibers veryinteresting and promising material for use in the strengthening of bio-polymer composites [22].

In this study, the dispersion ability of carrot nanofibers with differ-ent concentrations in polyol resin was studied. Furthermore, the effectof the CNF concentration for the foammorphology; cell size distribution,porosity and closed cell content were studied. The compressive me-chanical properties of the prepared foams were measured, and themodulus of the solid material in the foam was calculated. Finally, thefoam performance was compared with commercial PU foams.

2. Experimental

2.1. Materials

A castor oil polyol, Jagropol-400, Jayant Agro-Organics Ltd., India,with a hydroxyl value of 350 mg KOH/g, a viscosity between 700 and1100mPa·s, and a functionality of 3was used. A commercial isocyanate,

4,4′-polymeric methane diphenyl isocyanate, (ISO pMDI 92140) with afunctionality of 2.7, a –NCO content of 32% and a viscosity of 250mPa·s,was purchased from Lagotech AB, Sweden.

A catalyst, N,N-dimethylcyclohexylamine (DMCHA) was purchasedfrom VWR, Sweden. A polyether-modified polysiloxane surfactant,TEGOSTAB® B8433, was purchased from Evonik, Denmark. Distilledwater was used as a blowing agent.

Carrot nanofibers (0.5 wt% concentration in water) consisting ofmainly cellulose, were separated from bleached raw juice residue. Thepreparation of carrot nanofibers and their characteristics are describedelsewhere by Siqueira et al. [22].

2.2. CNF dispersion and nanocomposite foam preparation

2.2.1. CNF dispersion in polyol resinIn general, the water in CNF dispersion directly influences the

foaming process, and the aqueous dispersion of CNFs in polyol is diffi-cult. Therefore, a co-solvent, dioxane, was used to assist CNF dispersionin polyol resin. This co-solvent assisted method was carried out at in-creased temperature and high shear to form amutuallymiscible systemof water, CNFs, dioxane and polyol, fromwhich the volatiles (water anddioxane)were gradually removed, transferring the CNFs from thewaterphase to the polyol phase, to obtain well-dispersed CNFs in polyol. Inthis method, one third of the polyol and CNF suspension were mixedusing an Ultra-Turrax homogenizer at 10,000 rpm for 5 min. Thewater-CNF-polyol mixture was then transferred onto a hotplate (withmagnetic stirrer) set at a temperature of 70 °C and stirring at a constantspeed of 600 rpm. The water was slowly evaporated and gradually re-placed with dioxane. It was observed that in the presence of dioxane,which is miscible with both water and polyol, a relatively transparentmixture is formed at high temperature and shear stirring. This transpar-ency implies that the components of system are compatible with eachother, and the high temperature and shearing ensured that the CNFsremained dispersed even when the volatiles were removed. After 2 h,the remaining polyol was added in three equal portions at intervals of1 h. During this period, the dioxane added was twice the amount ofthe polyol. Then, the temperature was increased to 90 °C to evaporatethe remainingwater and dioxane from themixture under stirring. Final-ly, a transparent dispersion was achieved, indicating good dispersion ofthe CNFs in the polyol, whichwas confirmed by opticalmicroscopy. ThisCNF dispersion was used to prepare nanocomposite foams.

2.2.2. Preparation of nanocomposite foamsThe basic formulation of the BPU foamswas castor oil polyol, catalyst

1 phr, surfactant 2 phr and water as blowing agent 3.5 phr. The amountof different components in phr (per hundred resin) is based on thepolyol weight. To this basic formulation, dispersions of CNF at differentconcentrations (0.25, 0.5 and 1 phr) were mixed using mechanical stir-ring at room temperature for 30 s. Then, the pMDI was added andstirred for a further 20 s. The neat BPU and BPU-nano-compositefoams were obtained within a few seconds. The foams were removedfrom the mold after 30 min and cured under ambient room conditionsfor 1 week prior to sample preparation. The pMDI index (moles ofNCO groups/mol of OH groups) was adjusted to 1 for each of the foams.

2.3. Characterization

The dispersion of the CNFs in the polyol mixture was studied using apolarized optical microscope, Nikon Eclipse LV 100V (Japan). The vis-cosity of the CNF and polyol resin mixture was measured at 25 °Cusing a Vibro viscometer SV-100 (0.3–10,000 mPa·s), Japan.

2.3.1. Cell morphologyThe microstructure of the foams, parallel to the rising direction, was

studied under a scanning electron microscope (SEM) (JEOL JSM-IT300LV, Japan) using an acceleration voltage of 10 kV. Foams were

528 X. Zhou et al. / Materials and Design 110 (2016) 526–531

fractured after immersion in liquid nitrogen for 1 min, and the resultingcross-sections were sputter coated with gold to avoid charging. The cellsizes were manually measured using SEMAfore image analysis anddetermined by the Feret diameter. For each material, the average cellsize was calculated from N100 measurements.

2.3.2. Bulk densityThe densities of the foams in kg/m3 were obtained as the ratio of

weight to (geometric volume) of a cubic specimen with dimensions30 × 30 × 30 mm, according to ASTM D1622–08.

2.3.3. Percentage of closed cell contentA pycnometer, AccuPyc II 1340 (10 cm3 size sample chamber), with

nitrogen gas was used to determine the percentage of open cell content(Co) in the foams, according to a modified version of ASTM D2856 pro-vided by Micromeritics [23](Application Note 93). The density of solidpolyurethane (ρs) was obtained from a solid polyurethane preparationwithout foaming agent, and then the percentage of solid (cell wall andcell strut) volume occupied in BPU foam was calculated using the fol-lowing expression:

Sv ¼ m= ρsvð Þ½ � 100; ð1Þ

where m is the mass of the foam specimen, and v is the geometric vol-ume of the related foam specimen. Thus, the percentage of closed cellcontent was calculated using the following equation:

Cc ¼ 100−Co−Sv ð2Þ

Fig. 1. Polarized optical microscopy images of CNF dispersion in polyol resin (light phase CN0.25 phr CNF with co-solvent; c) polyol and 0.5 phr CNF with co-solvent; d) polyol and 1 phr C

2.3.4. Porosity measurementThe porosity (Φ) was calculated from the bulk density (ρb) of the

foam and the solid polyurethane density (ρs), as follows:

Φ ¼ 1− ρb=ρsð Þ ð3Þ

2.3.5. Mechanical propertiesThe compressive properties of the foams, were measured using a

universal testing machine (Instron 4411) equipped with a 5 kN loadcell and a constant load rate of 2.5 mm/min. The specimens were cutinto 51 × 51 × 26mmpieces, with the thickness parallel to the foam ris-ing direction. All samples were conditioned at 21 °C and 45% RH for atleast 40 h prior to the testing. All of the tests were performed accordingto ASTMD1621. Allmeasurementswere performed parallel to the risingdirection of the foam.

3. Results and discussion

3.1. CNF dispersion in polyol resin

Polarized optical microscopy was used to study the dispersion of theCNF in the polyol resin with and without dioxane co-solvent; the lightphase is carrot nanofibers. Fig. 1a shows the mixture with some largeCNF aggregates and Fig. 1 c the samemixture with the co-solvent, indi-cating successful CNF dispersion in the polyol. Fig. 1b shows themixturewith 0.25 phr, and 1 d shows the mixture with 1 phr. The mixture with1 phr is not as well dispersed as the lower-concentration mixtures,showing some nanofiber aggregation but not at same level as in theFig. 1a.

The co-solvent assisted method is demonstrated to be able to dis-perse the hydrophilic carrot nanofibers well into the hydrophobic

F) a) polyol and 0.5 phr CNF without co-solvent, showing CNF aggregates; b) polyol andNF with co-solvent.

Fig. 2. Foam appearance (a) and the microstructure of neat BPU (b) and BPU foam with 0.25 phr CNF (c), showing that the porosity and pore size are very similar.

529X. Zhou et al. / Materials and Design 110 (2016) 526–531

polyol. The increased dispersion of the CNFs in the polyol influencedthe shear resistance force. The viscosity was slightly increased with0.5 phr CNF and much more dramatically increased with 1 phr CNF,from 1087 mPa·s for the polyol resin to 6102 mPa·s for polyolresin with 1 phr CNF, which can hinder the foaming. Prolongo et al.[24] reported similar tendency in a previous study, in which carbonnanotubes and carbon nanofibers were dispersed in an epoxy resin:a mixture with a higher viscosity was obtained,

3.2. Behavior of CNF-reinforced BPU foams

The appearance and microstructure of prepared BPU foam andCNF-reinforced BPU nanocomposite foams are shown in Fig. 2. Allof the foams are white and uniform, with no obvious visual differ-ences due to the addition of CNFs. The morphology and cellularshape of the foams with increasing CNF content remain very similar,as shown in Fig. 2. Most of the cells are closed cells, and of sphericalshape. Some cells in these images are brighter because of the foldedcell edge.

Fig. 3 shows more detailed images of the cell struts and cell wallsof the foams. The cell struts and cell walls of nanocomposite foamswith low proportions of CNF (0.25 and 0.5 phr) are very similar tothe neat BPU foam, indicating that these low CNF contents are veryuniformly dispersed and show good interaction with the BPU poly-mer, compared with 1 phr CNF, where some dots and projections

Fig. 3. Cell strut and cell wall images of the prepared foams under highmagnification SEM, and twith 1 phr CNF.

become detectable on the cell wall and a coarse surface can be seenin the cell struts. These features can be attributed to CNFs embeddedin the cell struts and cell walls and are believed to be aggregates. Ingeneral, the good dispersion of nanofibers in polyol resin and theuniform distribution of nanofibers in the polymer matrix could playimportant roles in improving the compressive strength and modulusof the resulting nanocomposite foams, as will be discussed in detaillater. The cell strut area and Feret diameter distributions of BPUfoams with and without CNF are illustrated in Fig. 3.

The cell size distribution exhibits a slightly smaller range with in-creasing CNF content, resulting in a more uniform foam and smallercell size compared to the neat BPU foam. This feature is also confirmedby the SEM images in Figs. 2 and 3. It is possible that the CNFs act as nu-cleating agents for the cells and therefore have a positive effect on uni-form cell size distribution and decreased cell size.

3.3. Density, porosity and closed cell content

The densities of the nanocomposite foams are summarized inTable 1. No significant changes in the foam density are observed withthe addition of the CNFs.

The porosity and closed cell content are also listed in Table 1. Theporosity is approximately 96% for all of the foams, and the closed cellcontent is almost 90%, except for the nanocomposite foam with0.5 phr CNF, which was slightly lower, approximately 79%.

he size distribution: a) BPU, b) BPUwith 0.25 phr CNF, c) BPUwith 0.5 phr CNF and d) BPU

Table 1Cell characteristics of neat BPU foam compared to nanocomposite foams with differentCNF contents.

CNF content(phr)

Bulk density(kg/m3)

Skeleton density(kg/m3)

Porosity(%)

Closed cellcontent (%)

0 43 ± 3 1110 ± 19 96 890.25 43 ± 2 1069 ± 31 96 870.5 46 ± 3 1018 ± 12 96 791 49 ± 3 1025 ± 36 95 88

Table 2Measured compressive foam modulus and estimated solid modulus of BPU foams withdifferent CNF contents.

CNF content(phr)

Compressive modulus of thefoam(MPa)

Estimated modulus of solidmaterial(MPa)

0 3.5 ± 0.7 3910.25 4.3 ± 0.6 4600.5 5.7 ± 0.9 5351 4.4 ± 0.9 386

530 X. Zhou et al. / Materials and Design 110 (2016) 526–531

3.4. Mechanical properties

The compressive strength and modulus of the prepared nanocom-posite foams are shown in Fig. 4. The compressive strength andmodulusof the neat BPU foamare 241 kPa and 3.5MPa, respectively. These prop-erties are very good if compared, for example, with our earlier resultswith palm oil based BPU foams strength 54 kPa and modulus 1 MPa[20]. The properties are further improved with the addition of CNF,the highest compressive strength and modulus was achieved at0.5 phr CNF content. The compressive strength increased from 241 to307 kPa and the modulus from 3.5 to 5.7 MPa. In addition, foam withhigher CNF content was more ductile, possibly caused by a fiber bridg-ing effect. However, when the proportion of CNF reached 1 phr, thecompressive strength and modulus slightly decreased, though remain-ing higher than the native BPU foam. This slight decrease is expectedto be due to aggregation of the CNF.

Siqueira et al. [22] demonstrated that nanopaper produced usingcarrot nanofibers exhibits excellent mechanical properties, i.e. a highelastic modulus of 13.3 GPa and high yield strength of 175MPa. In addi-tion, the size of cellulose nanofibers is small enough to strengthen thecell walls in the foams. The high specific surface area of cellulose nano-fibers also contributes to the composite properties because of the reac-tion between the –OH groups of cellulose nanofiber and the isocyanategroups of pMDI. To determine the effect of density on foam modulus,the following equation for the modulus evaluation is given by Gibsonand Ashby [25]. In closed cell polyurethane foam, the Young's modulusof the foam includes three contributing factors: edge bending, facestretching and gas compression.

E�Es

¼ ∅ρ�ρs

� �2

þ 1−∅ð Þ ρ�ρs

� �þ P0 1−2v�ð ÞEs 1−ρ�=ρsð Þ ð4Þ

where∅ðρ�ρsÞ2 and ð1−∅Þ ρ�ρs are the edge bending and face stretching, re-

spectively. P0ð1−2v�ÞEsð1−ρ�=ρsÞ is the gas compression, which is negligible for the

closed rigid foam. Thus, the above equation is simplified to

E�Es

¼ ∅ρ�ρs

� �2

þ 1−∅ð Þ ρ�ρs

� �ð5Þ

Fig. 4. Compressive stress-strain curves and modulus-strain curves of

where E⁎ and Es are themodulus of the foam and solid material, respec-tively; ρ⁎ and ρs are the bulk density of the foam and the density of thesolid material, respectively; and ϕ is the volume proportion of the solidmaterial in the f cell edges and (1- ϕ) is the remaining fraction that is inthe cell faces. ϕ is taken as 0.8, which is typical for rigid polyurethanefoams [26]. The calculatedmodulus values of the foam and skeletonma-terial (solid material) based on Eq. (5) are shown in Table 2.

Themodulus of the nanocomposite foams and the estimated modu-lus of the solid material increases significantly with increasing propor-tion of CNFs up to 0.5 phr. The estimated modulus of the solidmaterial increased by 37% for this material. This result indicates thatthe reinforcement of the foammodulus is not only due to the foamden-sity but also due to the increase in modulus of the solid material.

3.5. Performance of nanocomposite foams

Fig. 5 shows how themechanical properties of the foams of themostsuccessful CNF reinforced BPU foams produced in this study, the 0.5 phrCNF foam, compared to commercial PU foams.

The addition of the CNF brings both themodulus and the strength ofthe BPU based foams in line with the commercially available rigidfoams. Thus, when density is considered, these nanocomposite foamsshow a commercially viable performance with regards mechanicalproperties.

Their use in a sandwich structure was also investigated by measur-ing the flexural stiffness of the foams with and without a skin of paperimpregnated with epoxy (Fig. 6). The chart shows that the flexuralproperties of the foams are in line with those of rigid PU foams, as wasthe case for the compressive properties of the nanocomposite foam.

It also shows that sandwich structure can be successfully made fromthe foams and that as expected, enhances the flexural modulus.

4. Conclusions

Rigid biopolyurethane foamswere successfully reinforced by achiev-ing good dispersion of carrot nanofibers (CNF) in castor oil polyolfollowed by a free-rising foaming method with water as blowing agent.

the native BPU foam and reinforced BPU nanocomposite foams.

Fig. 6. a) Vacuum infused sandwich composite with CNF reinforced BPU core and cellulose skin layer, and b) a chart of the flexural modulus verse density of the prepared nanocompositefoam and its sandwich structure in comparison with other PU foam families. Data from CES Edupack 2015 software [27].

Density (kg/m^3)40 60 80 100 120 140 160 180 200

Co

mp

ress

ive

mo

du

lus

0,005

0,01

0,02

0,05

0,1

Addition of CNF

Rigid PU

Flexible PU

Rigid BPU

Density (kg/m^3)40 60 80 100 120 140 160 180 200

Co

mp

ress

ive

stre

ng

th(M

Pa)

0

0,5

1

1,5

Addition of CNF

Rigid PU

Flexible PU

Rigid BPU

ba

(MP

a)

Fig. 5. Charts of a) the compressive modulus and b) the compressive strength verse density for CNF reinforced BPU foams and other PU foam families. Data from CES Edupack 2015software [27].

531X. Zhou et al. / Materials and Design 110 (2016) 526–531

The addition CNFs up to 0.5 phr improved the compressive strengthand modulus of the foams with no significant effect on the density, po-rosity or closed cell content.

Estimations from the Gibson's equation showed that this improve-ment is due to increase of the solidmaterialmoduluswhen CNF is added.

The prepared BPUnanocomposite foams are very promising becausethismaterial has a high strengthwith a lower density compared to othermany other biobased PU foams and can also compete with commercialPU foams with similar density.

Acknowledgments

The authors would like to acknowledge the financial support underINCOM EC FP7, Grant agreement No: 608746.

References

[1] K. Badri, Z. Othman, S. Ahmad, Rigid polyurethane foams from oil palm resources, J.Mater. Sci. 39 (2004) 5541–5542.

[2] A. Guo, I. Javni, Z. Petrovic, Rigid polyurethane foams based on soybean oil, J. Appl.Polym. Sci. 77 (2000) 467–473.

[3] S.S. Narine, X. Kong, L. Bouzidi, P. Sporns, Physical properties of polyurethanes pro-duced from polyols from seed oils: II. Foams, J. Am. Oil Chem. Soc. 84 (2007) 65–72.

[4] R. Gu, S. Konar, M. Sain, Preparation and characterization of sustainable polyure-thane foams from soybean oils, J. Am. Oil Chem. Soc. 89 (2012) 2103–2111.

[5] A. Abdel Hakim, M. Nassar, A. Emam, M. Sultan, Preparation and characterization ofrigid polyurethane foam prepared from sugar-cane bagasse polyol, Mater. Chem.Phys. 129 (2011) 301–307.

[6] J. Merle, M. Birot, H. Deleuze, C. Mitterer, H. Carré, F. Charrier-El Bouhtoury, Newbiobased foams from wood byproducts, Mater. Des. 91 (2016) 186–192.

[7] N.V. Gama, B. Soares, C.S.R. Freire, R. Silva, C.P. Neto, A. Barros-Timmons, A. Ferreira,Materials Bio-based polyurethane foams toward applications beyond thermal insu-lation, Mater. Des. 76 (2015) 77–85.

[8] M. Zhu, S. Bandyopadhyay Ghosh, M. Khazabi, H. Cai, C. Correa, M. Sain, Reinforce-ment of soy polyol-based rigid polyurethane foams by cellulose microfibers andnanoclays, J. Appl. Polym. Sci. 124 (2012) 4702–4710.

[9] R. Gu, M.M. Sain, S.K. Konar, A feasibility study of polyurethane composite foamwith added hardwood pulp, Ind. Crop. Prod. 42 (2013) 273–279.

[10] M. Silva, J. Takahashi, D. Chaussy, M. Belgacem, G. Silva, Composites of rigid polyure-thane foam and cellulose fiber residue, J. Appl. Polym. Sci. 117 (2010) 3665–3672.

[11] M.I. Aranguren, I. Racz, N.E. Marcovich, Microfoams based on castor oil polyure-thanes and vegetable fibers, J. Appl. Polym. Sci. 105 (2007) 2791–2800.

[12] M.A. Mosiewicki, G.A. Dell'arciprete, M.I. Aranguren, N.E. Marcovich, PolyurethaneFoams Obtained from Castor Oil-based Polyol and Filled with Wood Flour, J.Compos. Mater. 43 (2009) 3057–3072.

[13] B. Dahlke, H. Larbig, H. Scherzer, R. Poltrock, Natural fiber reinforced foams based onrenewable resources for automotive interior applications, J. Cell. Plast. 34 (1998)361–379.

[14] A.F. Sousa, M. Matos, R.J.B. Pinto, C.S.R. Freire, A.J.D. Silvestre, One-pot synthesis ofbiofoams from castor oil and cellulose microfibers for energy absorption impactma-terials, Cellulose 21 (2014) 1723–1733.

[15] X. Luo, A. Mohanty, M. Misra, Lignin as a reactive reinforcing filler for water-blownrigid biofoam composites from soy oil-based polyurethane, Ind. Crop. Prod. 47(2013) 13–19.

[16] O. Faruk, M. Sain, R. Farnood, Y. Pan, H. H Xiao, Development of lignin andnanocellulose enhanced bio PU foams for automotive parts, J. Polym. Environ. 22(2014) 279–288.

[17] Y. Li, H. Ren, A.J. Ragauskas, Rigid polyurethane foam/cellulose whisker nanocom-posites: preparation, characterization, and properties, J. Nanosci. Nanotechnol. 11(2011) 6904–6911.

[18] Y. Li, A.J. Ragauskas, Ethanol organosolv lignin-based rigid polyurethane foam rein-forced with cellulose nanowhiskers, RSC Adv. 2 (2012) 3347–3351.

[19] M.A. Mosiewicki, P. Rojek, S. Michałowski, M.I. Aranguren, A. Prociak, Rapeseed oilbased polyurethane foams modified with glycerol and cellulose micro/nano- crys-tals, J. Appl. Polym. Sci. 132 (10) (2015) 41602.

[20] X. Zhou, M.M. Sain, K. Oksman, Semi-rigid biopolyurethane foams based on palm-oilpolyol and reinforced with cellulose nanocrystals, Comp. Part A 83 (2016) 56–62.

[21] A.I. Cordero, J.I. Amalvy, E. Fortunati, J.M. Kenny, L.M. Chiacchiarelli, The role ofnanocrystalline cellulose on the microstructure of foamed castor-oil polyurethanenanocomposites, Carbohydr. Polym. 134 (2015) 110–118.

[22] G. Siqueira, K. Oksman, S.K. Tadokoro, A.P. Mathew, Re-dispersible carrot nanofiberswith high mechanical properties and reinforcing capacity for use in composite ma-terials, Combust. Sci. Technol. 123 (2016) 49–56.

[23] Application Note 93, Determining the Open Cell Content of Rigid Cellular Plasticswith the AccuPyc 1330 Pycnometer, http://www.micromeritics.com/Reposity/Files/ap93.pdf.

[24] S. Prolongo, M. Gude, A. Ureña, Synthesis and characterisation of epoxy resins rein-forced with carbon nanotubes and nanofibers, J. Nanosci. Nanotechnol. 9 (2009)6181–6187.

[25] L.J. Gibson, M.F. Ashby, Cellular Solids: Structure and Properties, second ed. Cam-bridge University Press, 1997 192–197.

[26] D.W. Reitz, M.A. Schuetz, L.R. Glicksman, A basic study of aging of foam insulation, J.Cell. Plast. 20 (1984) 104–113.

[27] G. Design, CES Edupack Software - Level 3 Eco Design, Granta Design Ltd., Cam-bridge, 2015.