Nanostructured rigid polyurethane foams with improved specific thermo-mechanical properties using Bacterial Nanocellulose as a Hard Segment
Sofia Benavides, Franco Armanasco, Patricia Cerrutti, Leonel Matías Chiacchiarelli
NN ANOSTRUCTURED RIGID POLYURETHANE FOAMS WITHIMPROVED SPECIFIC THERMO - MECHANICAL PROPERTIES USING B ACTERIAL N ANOCELLULOSE AS A H ARD S EGMENT
A P
REPRINT
Leonel M. Chiacchiarelli ∗ Instituto de Tecnología de Polímeros y Nanotecnología (ITPN)CONICET-UBA,Buenos Aires, Argentina.Instituto Tecnológico de Buenos Aires,Departamento de Ingeniería Mecánica,Av. E. Madero 399, Buenos Aires, Argentina.
Sofia Benavides
Instituto de Tecnología de Polímeros y Nanotecnología (ITPN)CONICET-UBA,Buenos Aires, Argentina.
Franco Armanasco
Instituto de Tecnología de Polímeros y Nanotecnología (ITPN)CONICET-UBA,Buenos Aires, Argentina.Instituto Tecnológico de Buenos Aires,Departamento de Ingeniería Mecánica,Av. E. Madero 399, Buenos Aires, Argentina.
Patricia Cerrutti
Instituto de Tecnología de Polímeros y Nanotecnología (ITPN)CONICET-UBA,Buenos Aires, Argentina.Departamento de Ingeniería Química,Facultad de Ingeniería, UBA,Buenos Aires, Argentina.July 20, 2020 A BSTRACT
Bacterial nanocellulose (BNC) was used to synthesize rigid polyurethane foams (RPUFs) based onits reaction with the isocyanate precursor (ISO route) and also by using the conventional procedure(POL route). The results indicated that at only 0.1 wt. % of BNC, drastic improvements of specificelastic compressive modulus (+244.2 %) and strength (+77.5 %) were found. The reaction of BNCwith the precursor was corroborated through the measurement of isocyanate number and the BNCcaused a significant nucleation effect, decreasing the cell size up to 39.7%. DSC analysis revealedthat the BNC had a strong effect on post-cure enthalpy, decreasing its value when the ISO routewas implemented. DMA analysis revealed that the RPUFs developed using the ISO route proved ∗ [email protected], TEL/FAX: +541143043020, Av. Gral Las Heras 2214, AAR1127, Buenos Aires, Argentina a r X i v : . [ phy s i c s . a pp - ph ] J u l anostructured rigid polyurethane foams with improved specific thermo-mechanical properties using BacterialNanocellulose as a Hard Segment A P
REPRINT to have an improved damping factor, regardless of BNC concentration. These results emphasizethe importance of using the ISO route to achieve foamed nanocomposites with improved specificmechanical properties.
Keywords rigid polyurethane foams · bacterial nanocellulose · thermo-mechanical properties Rigid polyurethane foams (RPUFs) represent a material of choice when energy efficiency and industrialized buildingsystems are taken into account [1, 2]. Nowadays, RPUFs are widely applied for the construction of sandwich panelswith metallic faces and RPUF cores [3] as well as for spray-up of conventional building structures. Traditional RPUFformulations comprise the use of a polyol and an isocyanate. Both components are produced industrially mostlyfrom non-renewable resources. This aspect represents a drawback towards the sustainability of the industry. To thiseffect, several studies have focus on the development of bio-based polyols [4, 5, 6, 7, 8, 9] and very few on bio-basedisocyanates [10, 11, 12, 13, 14, 15, 16, 17, 18].The formulation of a RPUF contemplates the use of moderate to highly functional polyols with low molecu-lar weight and aromatic polymeric diisocyanates. The final structure of the former in the polyurethane is termed as SoftSegment (SS) whereas the last one is termed as the Hard Segment (HS). The molecular structure of both componentsinduces the formation of a highly crosslinked polymer where the HS has a main role on the final mechanical propertiesof the RPUF. This characteristic has a profound implication on how the RPUF is formulated. In effect, formulationswhich have a highly crosslinked nature will also have a higher weight percent of isocyanate precursors. A clearexample of this is the Rigid Polyisocyanurate Foam (RPIR), where a very high isocyanate index is used so as to induceisocyanate trimerization [19]. However, the main drawback of this strategy is that most of the research that has beenperformed in the development of bio-based polyurethane precursors is associated to the polyol component. Hence, amaterial with a higher renewable content will be difficult to attain because the main component of the RPUF is theisocyanate. To surmount this issue, another strategy can be based on the use of nanotechnology. Any nanoparticle (NP)introduced in the formulation which also has the role of a HS has a high potential to ameliorate the effect explainedabove. In addition, if those NPs are obtained from renewable resources, a RPUF with a higher renewable content willalso be achieved.As already explained by the author in a recent publication [20], several NPs have proven to produce substan-tial improvements of the thermo-mechanical properties of RPUFs. One of the main advantages of using NPs isthat its high intrinsic surface area implies that only a reduced amount of NP (< 1 wt.%) can have a substantialeffect on the final properties of the RPUFs. As a matter of fact, Chiacchiarelli et al.[21] demonstrated significantimprovements of the specific properties of RPUFs using only 0.2 wt.% of Bacterial Nanocellulose (BNC). Otherrecent studies have also emphasized on these aspects [22, 23], as well as the relevance of improving specificmechanical properties [24]. This final aspect should not be underestimated. The NPs have to improve specificproperties and the results should be reported taking into consideration the foam apparent density. Among all theNPs available both in industry and for research purposes, the ones derived from lignocellulosic sources representone of the most promising candidates [25, 26]. The main reasons have to do with its lower density, low ornon-existent toxicity, renewable nature and low cost. In fact, BNC is recognized as a GRAS material. In addition, itshould be noticed that BNC is a 2D nanoparticle or a nanofiber, where only the length is within the order of micrometers.Among those NPs, nanocellulose obtained both by hydrolysis (cellulose nanocrystals, CNC) and bacterialmeans (Bacterial Nanocellulose, BNC) are, nowadays, fields which are intensively studied [27, 28, 29]. However,very few studies have dealt with the development of RPUFs nanostructured with BNC, which is the main focus ofthis work. Gimenez et al. [21] found that the incorporation of only 0.2 wt.% in PUFs synthesized from castor oilcaused improvements in the thermo-mechanical properties. Recently, Chiacchiarelli et al. [30] studied the compressivemechanical behavior of RPUFs obtained from non-renewable resources and nanostructured with BNC. Other workshave also focus on BNC [31, 32, 33] but with solid polyurethane matrices.A key aspect towards the use of NPs in polymers has to do with how it is inserted and dispersed [34, 35, 36]. Most ofthe NPs are synthesized as aggregates, such as tactoids, and those have to be intensively mixed in order to achievea nanometric dispersion. In this regard, it is important to emphasize the conceptual distinction between dispersionprocedure and insertion route [35]. The latter one has to do with where the NP is dispersed or reacted, whereas theformer one has to do with the physical procedure used for the dispersion of the NPs establishing, a priori, a specificinsertion route. For example, if we change the dispersion method by using ultrasonic mixing instead of shear mixing,then, the dispersion procedure is being modified. On the other hand, if we disperse de NP in the isocyanate instead of2anostructured rigid polyurethane foams with improved specific thermo-mechanical properties using BacterialNanocellulose as a Hard Segment
A P
REPRINT the polyol, then, the insertion route is being modified.Most of the literature regarding nanostructured polyurethanes involves the dispersion of the NPs in the polyolcomponent [24, 37, 23, 38]. This insertion route (POL-ROUTE) is straightforward because no chemical reaction takesplace between the NP and the polyol. The NP remains in the polyol as a colloid, and the stability of this colloidaldispersion [39] is a key issue for the successful implementation of this route. Since most polyols are hydrophilic innature, only NPs with similar properties might be dispersed using this route. A disadvantage of this route is revealedwhen the nanostructured polyol reacts with the isocyanate component. In fact, it is not certain if the functional groupsof the NP will either react with it, forming additional HS or, instead, the NPs will remain as a SS. In this last case, theNP will not have a profound effect on the thermo-mechanical properties of the resulting material. In addition, the NPmight also affect physical crosslinking within the SS generated by the molecular structure of the polyol, having anadverse effect in mechanical properties. An alternative insertion route is related to the dispersion of the NP in theisocyanate precursor (ISO-ROUTE). In this case, both dispersion as well as chemical reactions are feasible or desirableoutcomes. Very few studies have focus on this route [40, 41], and as far as the authors are concerned, none has beenperformed in the field of RPUFs. The main advantage of this insertion route has to do with the fact that if the NPgenerates a chemical bond with the isocyanate, it will certainly act as a HS. The NP will be a part of the molecularstructure of the polymer before the final preparation of the RPUF through the reaction of the isocyanate and formulatedpolyol components. Until now, no other work has dealt with this route using BNC as the NP.This work will focus on the development of nanostructured RPUF using the POL and ISO routes. BNC wasincorporated with both routes in a concentration range of 0 to 0.5 wt.%. In-situ temperature rise experiments wereperformed to measure how the kinetics was affected by BNC content and route. Differential Scanning Calorimetry wasused to analyze the thermal transitions of the resulting nanostructured RPUFs. Further characterization techniquesinvolved fracture surface analysis with SEM, thermomechanical analysis using Dynamical Mechanical ThermalAnalysis (DMA), Isocyanate number (NCO value ) of the nanostructured isocyanate as a function of BNC content andchemical groups present in the polymer using FTIR.
A Polymeric Methylene Diphenyldiisocyanate (PMDI, Suprasec 5005) was used as received. It had a functionality of2.70 and a NCO number of 31.0. Before each experiment, the NCO number of the isocyanate batch was measured (ASTMD2572) so as to corroborate that relevant changes on the NCO number did not occur. A commercially available polyetherpolyol was used (Rubitherm LP18497). The physical blowing agent was the HCFC141. The formulation of the polyolcomponent consisted of 100 parts by weight (pbw) of Rubitherm LP18497, 2 pbw of Dimethylcyclohexylamine(Rubitherm 18412) and 14 pbw of HCFC141. It is important to notice that this blowing agent is being phased out due toits high ozone depletion potential (ODP). Future research will involve the use of BAs with lower ODP. Finally, theisocyanate component was mixed at 160 pbw considering a 100 pbw basis of Rubitherm LP18497, giving an isocyanateindex (NCO index ) of 1.05. Measurement of NCO number was achieved following the ASTM D2572 standard using drytoluene (Biopack, A.C.S. grade), diisobutylamine (99%, Sigma-Aldrich), isopropyl alcohol (Biopack, A.C.S. grade),bromphenol blue indicator (Biopack) and HCl (Biopack, A.C.S. grade).
Bacterial nanocellulose (BNC) was produced by a strain of
Gluconacetobacter xylinus
NRRL B-42 gently provided byDr. Luis Lelpi (Fundación Instituto Leloir, Buenos Aires, Argentina). Static fermentations were carried on for 14days in Hestrin and Schramm medium [42] modified by replacing D-glucose by the same concentration of glycerol(Biopack) at 28+/-1° C , maintaining a ratio “volume flask: volume medium” of 5:1. The pellicles of BNC, from nowon denominated BNC mat, were rinsed with water to remove the culture medium and then boiled at 100° C in 2% w/vNaOH solution for 1 h in order to eliminate the bacterial cells from the cellulose matrix. Finally, the BNC mat waswashed with distilled water till neutralization. 3anostructured rigid polyurethane foams with improved specific thermo-mechanical properties using BacterialNanocellulose as a Hard Segment
A P
REPRINT
A scheme of the preparation procedures of the RPUFs synthesized in this work is depicted in Fig.1 . Both insertionroutes started from the BNC mat , which was composed by approximately 98 wt. % of water. The next step consisted inthe removal of the aqueous phase using lyophilization. A Labconco Freezone 2.5 equipment was set up with samples ofapproximately 15 g. and the process lasted 48 hs. at . × − bar. Then, the anhydrous mat was introduced into ahigh shear mixing device (Sparmix), obtaining microparticles suitable for dispersion in viscous liquids (BNC micro ).The POL route consisted on adding BNC micro in the polyol at concentrations of 0.1, 0.2 and 0.3 wt. % (theconcentrations were taken with respect to the total mass of the formulation). Then, the mixture was homogenized(Proscientific) for a total time of 5’ doing steps of 1’ so as to avoid excessive heat generation. Then, the blowing agentand the amine catalyst were added and the whole mixture was homogenized for 1’, obtaining a formulated polyol readyto react with the isocyanate.The ISO route consisted of adding the BNC micro into a glass reactor (Velp Scientific) with isocyanate atconcentrations of 0.1, 0.3 and 0.5 wt. %. Then, the temperature of the reactor was raised to 60°C while maintaining avigorous stirring with a magnetic bar. To avoid undesired reactions of isocyanate with water, the reactor was kept undera nitrogen atmosphere at all times. Before each experiment, the isocyanate was degassed using a dispermat LC30 mixerequipped with a vacuum system. After a total reaction time of 4 hours, the isocyanate prepolymer (isocyanate andBNC) was ready to react with the formulated polyol.Finally, for both routes, the formulated polyol and the isocyanate were poured in a HDPE cylindrical mouldwith an internal diameter of 70 mm and dispersed with a Cowles stirrer rotating at 2000 rpm for 30 seconds. The in-situtemperature evolution of the foaming process was measured with a K type thermocouple connected to a data logger(TES 1307). To ensure reproducibility, three samples were synthesized for each formulation studied.The BNC concentration denoted in this work corresponded to the total weight of the polyurethane formula-tion and not with respect to the polyol or the isocyanate components. Differential Scanning Calorimetry (DSC) was performed with a Shimadzu DSC-60.To obtain the thermal transitions ofthe formulations, a full thermal cycle was implemented which consisted of three thermal cycles. The first one started at25° C and went up to 150 ° C at a scan rate of 5 ° C min − . The second one started at 200° C and went down to 25° Cat 25° C min − and, finally, the third one was identical to the first cycle.Compressive mechanical analysis was performed with an Instron 5985 universal testing machine, followingthe guidelines of the standard ASTM D1621. For each formulation, 15 samples were tested from three different foamsprepared at identical conditions. Apparent density was calculated by measuring the sample dimensions and weight.FTIR absorption spectra were obtained with a Shimadzu IRAffinity using the absorption methodology. Thinsheets of 0.2 mm were extracted from the foam samples and those were used directly in the apparatus. Due to theporosity of the samples, the spectra were recorded without a significant attenuation loss. It were obtained from 40 scansat a resolution of 4.0 cm − and using the Happ-Genzel apodization function incorporated in the IRAffinity softwareincorporated in the FTIR device. The absorption spectra were normalized using the band of the phenyl group, centeredat approximately 1595 cm − .Dynamic Mechanical Analysis was carried out with a Perkin & Elmer DMA 8000. The flexural mode wasused, fixing the oscillation frequency to 1.0 Hz and the amplitude to 0.01 mm. The thermal scan started at 9° C andwent up to 180° C at a scan rate of 2° C min − . It was corroborated that the experiments fell into the linear viscoelasticregion of the material. At least three samples were tested for each formulation. Sample dimensions were typically of 10mm in length by 10.0 mm in width and 4.6 mm in thickness. Normalization of the results consisted in dividing thespecific elastic modulus by the apparent density of each sample tested.Regarding nomenclature, RPUF will refer generally to a Rigid Polyurethane Foam. If the POL or ISO routeare mentioned, then, the discussion will focus on a RPUF prepared using either route. When BNC is used in theformulation, then, the foam will be termed RPUF – x wt. % BNC, whereas the x represents de BNC concentration value.4anostructured rigid polyurethane foams with improved specific thermo-mechanical properties using BacterialNanocellulose as a Hard Segment A P
REPRINT -situ temperature rise, apparent density and cell size
The synthesis of a RPUF involves controlling chemical and physical phenomena taking place after mixing theformulated polyol and isocyanate precursors [1]. Any change in the formulation might modify the blowing or gellingkinetics to such an extent that the stages of cell formation and stabilization could be altered significantly [1]. Forexample, if the formulation causes a significant increase of temperature during rise, a friable or, in the worst scenario, aburnt core might be obtained. To this effect, an in-situ thermocouple is useful to understand if the formulation inducessignificant changes in either the gelling or the blowing reaction kinetics. In literature [23] it is usual to report creamtime, gel time, rise time and tack-free time. However, in this work we will focus mostly on the slope of the temperatureexcursion as well as the peak temperature rise during the foaming process.The in-situ temperature excursion, which is the in-situ temperature rise using the initial experiment tempera-ture as baseline for the foams synthesized in this work is depicted in Fig.2. To avoid confusion, representative data ispresented. The original data can be consulted by the reader in the journal supplementary information. For the case ofthe RPUFPOL, the average maximum temperature excursion (T max ) was 28.3 ° C. The effect of introducing BNC inthe formulation caused a monotonous decrease of the T max , reaching an average of 18.5° C at a BNC concentration of0.3 wt. %. Such result indicated that the BNC was changing the kinetics of either the gelling or the blowing reaction,competing to react with isocyanate groups and, thus, causing a decrease of the maximum temperature excursion. Thisbehavior is similar to what the author has found in other PUF formulations [20, 21, 43]. On the other hand, as it can bededuced from Fig.2, this tendency was not found for the ISO-route. In fact, the T max decreased to a less extent (8.1° C)and it was not dependent of BNC concentration. Such measurement was expected because the BNC reacted previouslywith the isocyanate, leaving only isocyanate groups to further react with the hydroxides groups of the polyol. These lastresults indicated another advantage of using the ISO-route instead of the POL-route. If no changes in reactivity arefound, then, it is not necessary to change the formulation for a specific application.The apparent density of the RPUF developed in this work using both the
ISO and
POL routes is reported inTab.1. The baseline RPUF pol had an apparent density of 46.4 +/- 4.7 Kg.m − . It showed negligible changes as afunction of BNC concentration, having a slight increase of its value (+5.6%) at the highest BNC concentration used forthis route (0.3 wt. %). Instead, the baseline RPUF iso had an apparent density of 40.5 +/- 3.4 Kg.m − . Its variation as afunction of BNC concentration was also negligible, except for the highest BNC concentration (0.5 wt. %), whereas theapparent density decreased by 8.4%.The cell size measured in both the longitudinal growth direction (L) and the transverse one (T), as well asthe anisotropy factor (AF), are reported in Tab.1. As it can be easily noticed, the L direction cell size was significantlyhigher than the T direction in all the cases analyzed. Such result was expected due to the fact that the foam growthwas constrained in the T direction by the mould walls. For the baseline RPUF of the POL-route, the cell size in the Ldirection was . × µ m. This dimension decreased as a function of increasing BNC concentration (except only forthe case of 0.2 wt. %), having a maximum decrease at 0.3 wt. % of -16.7%. This indicated that, from a cell size point ofview, the BNC acted as a nucleation agent. For the baseline RPUF of the ISO-route, the cell size in the L direction was . × µ m. In analogy with the other route, the cell size decreased as a function of increasing BNC concentration,but to a greater extent. The maximum decrease, -39.7%, was measured at a BNC concentration of 0.3 wt. %. Finally,the AF measured for both the ISO and POL routes did not change significantly, even as a function of BNC content. The isocyanate number (NCO number ) is a measure of how many free isocyanate groups are available to react withthe formulated polyol. The NCO number should not be confused with the isocyanate index (NCO index ), which isassociated to the molar excess of isocyanate groups present in the formulated system. As already stated in the materialssection, the isocyanate used in this work was a polymeric methylene diphenyldiisocyanate (pMDI). Due to the fact thatMDI is solid at room temperature, it is usual to synthesize an uretoneimine modified MDI through the carbodiimideintermediate route [44]. Such route enables the formation of an eutectic, which is liquid at room temperature. Thisvariant is nowadays used as a standard in both industrial as well as scientific studies.The selection of the insertion route of BNC in the polyurethane system has a profound effect on the finalproperties of the RPUF [35]. The POL route involves the insertion of the nanoparticle in the polyol component. In thisscenario, no chemical reaction can take place and the formation of a nanometric dispersion is the most relevant issue so5anostructured rigid polyurethane foams with improved specific thermo-mechanical properties using BacterialNanocellulose as a Hard Segment
A P
REPRINT as to obtain nanocomposites with improved mechanical properties. However, such strategy imposes a limitation on thetype of nanoparticle to be inserted. Due to the fact that most polyols are hydrophilic, the nanoparticles should also havea similar hydrophilic character. Otherwise, the colloidal dispersion will not be stable and the nanoparticle will forma denser second phase, with a high probability of gravity induced precipitation. In addition, the nanoparticle is notchemically linked with the polyol and, if it has an hydroxide functionality, it will compete with the isocyanate freeradicals to form chemical linkages. The outcome of that competition will dictate if the nanoparticle will behave aseither a SS or a HS. A final consideration has to do with the rheology of the dispersion. If significant changes of therheological response as a function of shear rate are encountered, the polyol might not be able to properly mix with theisocyanate component.On the other hand, the nanoparticle can be inserted using the ISOMIX route [35]. In this work, the proce-dure is termed ISO route. The main objective is to generate a chemical linkage with the isocyanate monomers, reducingthe overall NCO number . Such a route is frequently used in industry to obtain elastomeric polyurethanes. However, it isnot frequently employed to develop polyurethane nanocomposites. The main advantage of this route has to do with thefact that no second phase is formed, hence, precipitation and dispersion stability do not take place. In addition, thechemical linkage formed by the surface functionality of the nanoparticle and the free isocyanate group ensures that thenanoparticle will behave as a HS. However, the main disadvantage of this route has to do with the fact that the reactionof the isocyanate and the nanoparticle has to be performed in an inert atmosphere, otherwise, undesirable side reactionsmight take place. Finally, since the NCO number is being changed, additional measurements of the NCO number shouldbe performed, so as to have the correct free isocyanate groups that will react with the polyol.The NCO number as a function of BNC content is depicted in Fig.3 It is important to highlight that a blankrun was performed so as to evaluate the effect of the thermal protocol without the addition of BNC. The NCO number measured of the blank was termed as (ISO)T (Fig.3). As it can be noticed from this figure, the thermal protocol causeda reduction of the NCO number by 3.4 %. Such a reduction was expected due to the change of the eutectic equilibriumof the uretoniemine associated to a thermal treatment above 40°C. Further details about this phenomena were studiedby Hatchett et al. [44]. On the other hand, the addition of BNC in the thermal protocol caused a monotonous decreaseof the NCOnumber as a function of increasing BNC concentration, reaching a maximum reduction of 20.2 % of theNCOnumber for BNC concentrations of 0.5 wt. % (reaching a final NCO number of 23.7 %). This result is a strongsupport to the hypothesis that the functional groups present in the BNC are reacting with the isocyanate precursor.Further support of this hypothesis is presented in the FTIR section, whereas additional chemical links are found whenthe BNC is dispersed using the ISO route.
The absorption spectra of the RPUF obtained from the ISO and pol routes are depicted in Fig.4 The absorption band ofthe phenyl group (1595 cm − ) was used to normalize the spectra. The absorption bands of the urethane functionalitywere found at 1220 cm − and 1700 cm − . In addition, the isocyanurate and the urea groups were centered at 1410cm − and 1510 cm − , respectively. The isocyanate group was centered at 2270 cm − and the CO (g) molecule werealso found at 2341 cm − .Along with the foaming reaction, isocyanate reacts with water, releasing CO (g). This gas is purposely con-strained within the cells of the RPUF so as to reduce thermal conductivity. Hence, it was logical to find the CO (g)absorption band in the spectra and no attempt was made to perform an atmospheric correction of the spectra, becausethe CO (g) concentration within the cells was much higher than the one in the atmosphere of the FTIR instrumentby itself. Hypothetically, the CO (g) peak height can be correlated to specific foaming parameters during synthesis.Indeed, it is possible to argue that a higher height might imply that the isocyanate reacted more effectively with water,releasing a higher amount of CO (g) within the cell and having a profound impact on the thermal conductivity of theRPUF. However, such attempt to quantify the spectra will only be valid if the outward diffusion of CO (g) is also takeninto consideration. Due to the fact that the main focus of this work was not the measurement of thermal conductivity asa function of time (ageing), such analysis will not be further discussed.An interesting aspect of the spectra depicted in Fig.4 has to do with the isocyanate absorption band. In thisregard, it is important to highlight that the occurrence of such band reflected the fact that unreacted isocyanate waspresent after the foaming reaction took place. The first important aspect which needs to be noticed is that the FTIRspectra depicted in FTIR represented samples which were cured at ambient temperature. The second aspect whichwas important was the variation of peak height as a function of route and BNC concentration. As it can be noticed,the route had a relevant effect on the height of the isocyanate band. In effect, it can be deduced for the POL route,6anostructured rigid polyurethane foams with improved specific thermo-mechanical properties using BacterialNanocellulose as a Hard Segment A P
REPRINT the height of the band was in general higher with respect to the ISO route. This effect is particularly noticeable at aBNC concentration of 0.3 wt. %, where the highest peak height was measured with respect to all the samples underanalysis. This analysis can support the hypothesis that BNC is changing substantially the cure kinetics of the RPUFobtained using the POL route. In contrast, the cure kinetics of the RPUF obtained from the ISO route were notaffected in this regard, because the BNC was previously reacted with the isocyanate component. In fact, the height ofthe isocyanate absorption band in all the spectra of the ISO route were small and also independent on BNC concentration.
As already noticed in the methodology section, DSC analyses were performed using a full thermal cycle. The Heat Flux(HF) as a function of Temperature of the RPUF for the first and second thermal cycle is depicted in Fig.5. As it can bededuced from Fig.5, the first cycle presented an exothermic event, which can be associated to the post-cure of the RPUF( ∆ Hc). Taking into account that the foams were cured at room temperature (25°C) and that the NCO index was 1.05, itwas logical to obtain such result. On the other hand, the second cycle presented no exotherm, showing only a glass tovitreous transitions (Tg). This analysis was extended to all the foams synthesized in this work. To avoid confusion, theresults were summarized in Tab.1. The enthalpy of the first thermal cycle as a function of processing route and BNCcontent is depicted in the first column of Tab.1. From these results it can be deduced that the processing route hadan important effect on the ∆ Hc, where the transition from the POL to the ISO route caused an average decrease ofapproximately 16.9 %. On the other hand, it is important to notice how the ∆ Hc changed as a function of BNC content.For the case of the ISO route, no significant changes of the ∆ Hc as a function of BNC content were measured. On theother hand, the POL route presented significant changes of ∆ Hc as a function of BNC content. These results supportthe previously stated hypothesis that the BNC had a significant role in the cure kinetics of the foam during synthesis. Infact, as it can be deduced from the results, the BNC inserted using the POL route caused a significant inhibition of curekinetics. In contrast, the ISO route presented a negligible effect.The results presented in the last paragraph are supported and corroborated by the other experiments performed inthis work. This has a profound impact on the development of polyurethane nanocomposites and, especially, in thedevelopment of RPUFs. In industry, RPUFs are applied by industrial methods where cure is commonly performed atambient temperature. In other words, it is not frequent to find a high temperature (>40° C) post-cure in industry. Thisclearly indicates that if nanocomposites are to be developed in this particular field, any substantial change in curekinetics should be avoided, because it would not be possible to counteract this effect by performing a post-cure. Takinginto account this essential aspect, it can be concluded that the use of the ISO route is a key aspect for the developmentof nanostructured RPUFs. If the POL route is used, the reaction kinetics would have to be increased by a change offormulation, which would probably require the use of a higher amount of stannous or amino catalysts. Such path wouldinevitable attempt negatively in regard of the toxicity of the final formulation.
The compressive true stress as a function of deformation for the RPUF pol , RPUF iso as well as the foams nanostructuredwith BNC are depicted in Fig.6a and Fig.6b. Only representative curves were depicted, so as to avoid confusion. As itcan be noted, the evolution of the stress-strain curves presented three stages, the initial linear-elastic one, the subsequentplateau regime and the final densification stage [20]. For all the cases studied in this work, the stress-strain responsewas only measured in the longitudinal growth direction. This, in turn, meant that all the curves presented yield and,following the guidelines of ASTM 1621, that point was considered as the compressive strength of the foam. A studyof the stress-strain response of these foams in both the longitudinal and the perpendicular growth direction can beconsulted in a previous published work [20]. As well as the stress-strain curves, the values of specific compressivestrength and elastic modulus of the foams are also depicted in Fig.6c and Fig.6d.As it can be deduced from Fig.6, both the insertion route as well as the BNC had profound effects on themechanical properties of the RPUFs. Let’s focus first on the specific compressive strength (R csp ). As a remainder,the RCSP was calculated by normalizing the compressive strength with respect to apparent density. For the caseof the POL route, the BNC caused both a deterioration of the R csp by -8.7 % for a BNC concentration of 0.1 wt.% as well as an outstanding improvement of +53.6 % at a BNC concentration of 0.3 wt. %. This indicated thatsmall concentrations of BNC using the POL route were not effective to cause an improvement of the R csp . Only forconcentrations above 0.2 wt. % the R csp was improved. On the contrary, for the case of the ISO route, the highestimprovement of the RCSP, which was +77.5 %, was measured at the lowest BNC concentration (0.1 wt. %). A higherBNC concentration caused a deterioration of the improvement of the SCS and for the case of BNC loaded at 0.57anostructured rigid polyurethane foams with improved specific thermo-mechanical properties using BacterialNanocellulose as a Hard Segment
A P
REPRINT wt. %, the overall R csp value showed a deterioration of -4.00 %. A similar trend was also measured for the specificelastic modulus (E sp ) of the foams. For the case of the POL route, the BNC caused an improvement of the E sp as afunction of increasing BNC content, reaching a maximum improvement of +197.3 % at a BNC loading of 0.3 wt. %.For the case of the ISO route, a maximum improvement of +244.2 % was measured for a BNC loading of only 0.1 wt. %.Another relevant aspect that needs to be addressed from FIGc and FIGd is heterogeneity. As it can be de-duced from the results, a relevant statistical scattering was measured for the foams prepared with BNC. Forexample, for the case of the E sp of the RPUF prepared with the ISO route at a 0.1 wt. %, the scattering measuredwas in the order of 22%. Due to the fact that for each formulation we have synthesized three identical foamsand measured its properties using a total amount of 15 samples for each formulation, we can be certain that theintroduction of BNC in the system was the main cause of the measured scattering. Then, we hypothesize that suchscattering can be explained by the fact that the results are very sensitive to foam growth orientation. As alreadymeasured in a previous publication of our group [20], BNC caused a significant change of the direction wherethe cells aligned. The usual hypothesis that the growth direction coincides with the main axis of the cylindricalcup used in the foaming experiments was no longer valid and, in consequence, the results showed an increased scattering.The previously stated results have profound implications towards the selection of a specific route of insertionof the BNC. As it can be deduced, the POL route requires higher concentrations of BNC so as to measure significantimprovements in mechanical properties. On the other hand, the ISO route offered significant improvements only at thelowest BNC concentration (0.1 wt. %).It is important to highlight that the selection of the optimal insertion route does not only take into considera-tion the weight content of BNC in the formulation. The effect of BNC on viscosity and reactivity should also be takeninto account. For example, for the case of the POL route, higher concentrations were necessary to obtain mechanicalimprovements. However, this had a detriment effect on the polyol viscosity. A significant increase of viscosity will notrender the polyol suitable for a RPUF system. On the other hand, the ISO route caused significant improvements of themechanical properties for small concentrations, however, the dispersion process required the use of an inert atmosphereso as to avoid undesirable reactions of the isocyanate precursor.Unfortunately, we cannot compare our results with others in literature. This is the first publication regardingthe use of BNC in RPUFs. Our group recently published that the incorporation of BNC caused an improvementof specific mechanical properties of castor oil based polyurethane foams. Other works which have focus on CNC(crystalline nanocelluse), such as the one of Ragauskas et al. [45] and Septevani et al. [23] have found improvements ofmechanical properties, but the concentrations of CNC were higher and the resulting mechanical properties were notdramatically improved. The specific storage modulus (E’ sp ) and the damping factor (Tan delta) as a function of temperature for the RPUF pol ,RPUF iso as well as the foams nanostructured with BNC are depicted in Fig.7. To avoid confusion, only representativecurves are depicted. It is important to highlight that the foams were measured under flexural conditions. All themeasurements showed a similar behavior, with a localized thermal transition centered at approximately 30° C and abroad transition which was extended up to the final temperature of the experiment, which was 180°C. There resultswere expected because of the formulation of the polyol used in this work. Taking into account that two polyols wereused to formulate the polyol, then, it was logical to measure both thermal transitions.For the case of the POL route, the variation of the E’ sp as a function of temperature did not present a signif-icant variation as a function of BNC content. A slight increase of the slope of E’sp as a function of temperature wasmeasured, caused mainly for increasing BNC contents. The damping factor (tan δ ) consolidated this tendency, whereasthe maximum value of the damping factor decreased as a function of increasing BNC content. As far as the absolutevalue of E’ sp , for all BNC loadings the RPUF presented a higher E’ sp , with a maximum for the case of 0.1 wt. % BNC.This result might me understood as a contradiction with respect to the results found in the compression tests above.However, it is important to highlight that the stress states were different. DMA analyses were performed under flexuralconditions whereas quasi-static analysis were performed under compression. The differences found in this work are inagreement with what we have found in a previous work of our group [21]. Similar results were found for the case ofthe ISO route, except for the case of the damping factor. Indeed, the ISO route caused a relevant improvement of thedamping factor regardless of BNC concentration. 8anostructured rigid polyurethane foams with improved specific thermo-mechanical properties using BacterialNanocellulose as a Hard Segment A P
REPRINT
BNC nanostructured RPUFs were synthesized by the ISO and POL processing routes and using BNC concentrations ofup to 0.5 wt. %. The POL route contemplated the dispersion of BNC within the polyol component while the ISO routewas associated to a dispersion and reaction of the BNC in the isocyanate precursor. In-situ temperature measurementsindicated that the ISO route presented a lower T max (-10.2 ° C), indicating slower foam cure kinetics. The addition ofBNC in the formulation had a similar effect. Isocyanate NCO numbers were measured for all the samples studied.A monotonous decrease of the NCO number was measured as a function of increasing BNC concentration (ISOroute), indicating that the BNC was effectively reacting with the isocyanate precursor. Micrographic cell size analysissupported the hypothesis that the BNC had a profound nucleation effect, more pronounced in the ISO route (-39.7%)but also present in the POL route (-16.7%). No significant changes of apparent density were measured as a functionof processing route or BNC content. Both the ISO and POL route presented outstanding improvements of specificcompressive stress (R csp ) and elastic compressive modulus E sp . The highest improvement of E sp (+244.2 %) and R csp (+77.5 %) were found for the ISO route at the lowest BNC concentration (0.1 wt. %). DSC analysis revealed that thePOL route had a significant impact on post-cure enthalpy, increasing its nominal value as a function of increasingBNC content. On the other hand, the ISO route presented no significant changes of the cure enthalpy. DMA analysisunder flexural conditions supported the hypothesis that the ISO route improved the damping factor, regardless of BNCconcentration.The previous results reinforce the hypothesis that, for the case of RPUFs, the most suitable processing pathis the ISO route. Further research in this area should focus using this approach. Future work under development in ourlabs will focus on the effect of BNC nanoparticles using RPUFs obtained from soybased-polyols. Acknowledgements
The author would like to thanks colleagues which indirectly contributed to this work, Matías Nonna and Ma-tias Ferreyra (Huntsman) and Nicolás Andrés Oyarzabal (ITBA).
References [1] Randall, D. & Lee, S. (2003).
The polyurethanes book . Wiley.[2] Szycher, M. (2013).
Szycher´s Handbook of Polyurethanes . Florida, NY: Taylor & Francis group LLC.[3] Davies, J. M. (2008).
Lightweight sandwich construction.
John Wiley & Sons.[4] Chian, K. & Gan, L. (1998). Development of a rigid polyurethane foam from palm oil.
Journal of Applied PolymerScience , 68(3), 509-515, Wiley Online Library.[5] Guo, A., Javni, I., & Petrovic, Z. (2000). Rigid polyurethane foams based on soybean oil.
Journal of AppliedPolymer Science , 77(2), 467-473.[6] Hu, Y. H., Gao, Y., Wang, D. N., Hu, C. P., Zu, S., Vanoverloop, L., & Randall, D. (2002). Rigid polyurethanefoam prepared from a rape seed oil based polyol.
Journal of Applied Polymer Science , 84(3), 591-597.[7] Tan, S., Abraham, T., Ference, D., & Macosko, C. W. (2011). Rigid polyurethane foams from a soybean oil-basedpolyol.
Polymer , 52(13), 2840-2846.[8] Tu, Y. C., Kiatsimkul, P., Suppes, G., & Hsieh, F. H. (2007). Physical properties of water-blown rigid polyurethanefoams from vegetable oil-based polyols.
Journal of Applied Polymer Science , 105(2), 453-459.[9] Veronese, V. B., Menger, R. K., Forte, M. M. d. C., & Petzhold, C. L. (2011). Rigid polyurethane foam based onmodified vegetable oil.
Journal of Applied Polymer Science , 120(1), 530-537.[10] Cawse, J. L., Stanford, J. L., & Still, R. H. (1984a). Polymers from renewable sources, 1. Diamines anddiisocyanates containing difurylalkane moieties.
Macromolecular Chemistry and Physics , 185(4), 697-707.[11] Cawse, J. L., Stanford, J. L., & Still, R. H. (1984b). Polymers from renewable sources, 2. Kinetics and polyurethaneformation from furan-based diisocynates.
Macromolecular Chemistry and Physics , 185(4), 709-723.[12] Cawse, J. L., Stanford, J. L., & Still, R. H. (1985). Novel precursors suitable for RIM polyurethane networks.
Polymer International , 17(2), 233-238.[13] Çaylı, G., & Küsefo˘glu, S. (2008). Biobased polyisocyanates from plant oil triglycerides: Synthesis, polymeriza-tion, and characterization.
Journal of Applied Polymer Science , 109(5), 2948-2955. doi: 10.1002/app.284019anostructured rigid polyurethane foams with improved specific thermo-mechanical properties using BacterialNanocellulose as a Hard Segment
A P
REPRINT [14] Hojabri, L., Kong, X., & Narine, S. S. (2009). Fatty Acid-Derived Diisocyanate and Biobased PolyurethaneProduced from Vegetable Oil: Synthesis, Polymerization, and Characterization.
Biomacromolecules , 10(4), 884-891.doi: 10.1021/bm801411w[15] Hojabri, L., Kong, X., & Narine, S. S. (2010). Novel long chain unsaturated diisocyanate from fatty acid:Synthesis, characterization, and application in bio-based polyurethane.
Journal of Polymer Science Part A: PolymerChemistry , 48(15), 3302-3310. doi: 10.1002/pola.24114[16] Holfinger, M. S., Conner, A. H., Lorenz, L. F., & Hill, C. G. (1993). Difurfuryl diisocyanates: New adhesivesderived from renewable resources.
Journal of Applied Polymer Science , 49(2), 337-344.[17] More, A. S., Lebarbé, T., Maisonneuve, L., Gadenne, B., Alfos, C., & Cramail, H. (2013). Novel fatty acid baseddi-isocyanates towards the synthesis of thermoplastic polyurethanes.
European Polymer Journal , 49(4), 823-833.doi: http://dx.doi.org/10.1016/j.eurpolymj.2012.12.013[18] Zenner, M. D., Xia, Y., Chen, J. S., & Kessler, M. R. (2013). Polyurethanes from Isosorbide-Based Diisocyanates.
ChemSusChem , 6(7), 1182-1185.[19] Ashida, K. (2006).
Polyurethane and related foams: chemistry and technology:
CRC press.[20] Chiacchiarelli, L. (2019). ustainable, nanostructured, and bio-based polyurethanes for energy-efficient sandwichstructures applied to the construction industry.
Biomass, Biopolymer-Based Materials, and Bioenergy (pp. 135-160):Elsevier.[21] Gimenez, R. B., Leonardi, L., Cerrutti, P., Amalvy, J., & Chiacchiarelli, L. M. (2017). Improved specificthermomechanical properties of polyurethane nanocomposite foams based on castor oil and bacterial nanocellulose.
Journal of Applied Polymer Science , 134(25).[22] Andersons, J., Kirpluks, M., & Cabulis, U. (2020).
Modelling the Strength of Cellulose Nanofiber-Filled RigidLow-Density PU Foams.
Paper presented at the Key Engineering Materials.[23] Septevani, A. A., Evans, D. A. C., Annamalai, P. K., & Martin, D. J. (2017). The use of cellulose nanocrystalsto enhance the thermal insulation properties and sustainability of rigid polyurethane foam.
Industrial Crops andProducts , 107, 114-121. doi: https://doi.org/10.1016/j.indcrop.2017.05.039[24] Lobos, J., & Velankar, S. (2016). How much do nanoparticle fillers improve the modulus and strength of polymerfoams?
Journal of Cellular Plastics , 52(1), 57-88.[25] Brinchi, L., Cotana, F., Fortunati, E., & Kenny, J. (2013). Production of nanocrystalline cellulose from lignocellu-losic biomass: technology and applications.
Carbohydrate Polymers , 94(1), 154-169.[26] Charreau, H., Cavallo, E., & Foresti, M. L. (2020). Patents involving nanocellulose: analysis of their evolutionsince 2010.
Carbohydrate Polymers , 116039.[27] Fortunati, E., Yang, W., Luzi, F., Kenny, J., Torre, L., & Puglia, D. (2016). Lignocellulosic nanostructures asreinforcement in extruded and solvent casted polymeric nanocomposites: An overview.
European Polymer Journal ,80, 295-316.[28] Kim, J.-H., Shim, B. S., Kim, H. S., Lee, Y.-J., Min, S.-K., Jang, D., . . . Kim, J. (2015). Review of nanocellulose forsustainable future materials.
International Journal of Precision Engineering and Manufacturing-Green Technology ,2(2), 197-213.[29] Lizundia, E., Puglia, D., Nguyen, T.-D., & Armentano, I. (2020). Cellulose nanocrystal based multifunctionalnanohybrids.
Progress in Materials Science , 100668. doi: https://doi.org/10.1016/j.pmatsci.2020.100668[30] Chiacchiarelli, L. M., Cerrutti, P., & Flores-Johnson, E. A. (2019). Compressive behavior of rigid polyurethanefoams nanostructured with bacterial nanocellulose at low and intermediate strain rates.
Journal of Applied PolymerScience , 48701.[31] Juntaro, J., Ummartyotin, S., Sain, M., & Manuspiya, H. (2012). Bacterial cellulose reinforced polyurethane-basedresin nanocomposite: A study of how ethanol and processing pressure affect physical, mechanical and dielectricproperties.
Carbohydrate Polymers , 87(4), 2464-2469. doi: http://dx.doi.org/10.1016/j.carbpol.2011.11.020[32] Pinto, E. R., Barud, H. S., Polito, W. L., Ribeiro, S. J., & Messaddeq, Y. (2013). Preparation and characterizationof the bacterial cellulose/polyurethane nanocomposites.
Journal of Thermal Analysis and Calorimetry , 114(2),549-555.[33] Seydibeyo˘glu, M. Ö., Misra, M., Mohanty, A., Blaker, J. J., Lee, K.-Y., Bismarck, A., & Kazemizadeh, M. (2013).Green polyurethane nanocomposites from soy polyol and bacterial cellulose.
Journal of Materials Science , 48(5),2167-2175.[34] Kadam, H., Bandyopadhyay-Ghosh, S., Malik, N., & Ghosh, S. (2019). Bio-based engineered nanocompositefoam with enhanced mechanical and thermal barrier properties.
Journal of Applied Polymer Science , 136(7), 47063.10anostructured rigid polyurethane foams with improved specific thermo-mechanical properties using BacterialNanocellulose as a Hard Segment
A P
REPRINT [35] Kenny, J. M., Torre, L., & Chiacchiarelli, L. M. (2015). The effect of processing routes on the thermal andmechanical properties of poly (urethane-isocyanurate) nanocomposites.
Journal of Applied Polymer Science ,132(45).[36] Okolieocha, C., Raps, D., Subramaniam, K., & Altstädt, V. (2015). Microcellular to nanocellular polymer foams:Progress (2004–2015) and future directions–A review.
European Polymer Journal , 73, 500-519.[37] Pauzi, N. N. P. N., Majid, R. A., Dzulkifli, M. H., & Yahya, M. Y. (2014). Development of rigid bio-basedpolyurethane foam reinforced with nanoclay.
Composites Part B: Engineering , 67, 521-526.[38] Zhou, X., Sain, M. M., & Oksman, K. (2016). Semi-rigid biopolyurethane foams based on palm-oil polyol andreinforced with cellulose nanocrystals.
Composites Part A: Applied Science and Manufacturing , 83, 56-62.[39] Chiacchiarelli L. M., Puri I., Puglia D., Kenny J. M., & Torre L. The relationship between nanosilica dispersiondegree and the tensile properties of polyurethane nanocomposites.
Submitted to Colloid and Polymer Science(Minor Revisions). [40] Habibi, Y. (2014). Key advances in the chemical modification of nanocelluloses.
Chemical Society Reviews , 43(5),1519-1542.[41] Rueda, L., d’Arlas, B. F., Zhou, Q., Berglund, L. A., Corcuera, M., Mondragon, I., & Eceiza, A. (2011). Isocyanate-rich cellulose nanocrystals and their selective insertion in elastomeric polyurethane.
Composites Science andTechnology , 71(16), 1953-1960.[42] Harikrishnan, G., Macosko, C. W., Choi, J. H., Bischof, J. C., & Singh, S. N. (2008). A simple transient methodfor measurement of thermal conductivity of rigid polyurethane foams.
Journal of Cellular Plastics , 44(6), 481-491.[43] Herrán, R., Amalvy, J. I., & Chiacchiarelli, L. M. (2019). Highly functional lactic acid ring-opened soybeanpolyols applied to rigid polyurethane foams.
Journal of Applied Polymer Science , 47959.[44] Hatchett, D. W., Kodippili, G., Kinyanjui, J. M., Benincasa, F., & Sapochak, L. (2005). FTIR analysis of thermallyprocessed PU foam.
Polymer Degradation and Stability , 87(3), 555-561.[45] Li, Y., Ren, H., & Ragauskas, A. J. (2010). Rigid polyurethane foam reinforced with cellulose whiskers: Synthesisand characterization.
Nano-Micro Letters , 2(2), 89-94.11anostructured rigid polyurethane foams with improved specific thermo-mechanical properties using BacterialNanocellulose as a Hard Segment
A P
REPRINT
Table 1: Geometrical and mechanical properties of the RPUFs.Formulation T peak (° C) ρ avg (kg.m − ) Cell Size( µ m),Anisotropy Factor E c . ρ − avg (J.g − ) σ c . ρ − avg (J.g − ) ∆ H c (J/g)RPUF P OL +/- 3.2 4.64. +/- 4.66 L:3.83. +/-8.20. T:3.04. +/-6.00. AF: 1.27 1.17. − +/-1.45. − − +/-6.03. − +/-2.3RPUF P OL -BNC 0.1 wt.% 5.81. +/- 3.2 4.67. +/- 4.80 L:3.59. +/-8.09. T:2.86. +/-6.43. AF: 1.26 1.27. − +/-2.17. − − +/-9.40. − +/-2.5RPUF P OL -BNC 0.2 wt.% 4.55. +/- 1.4 4.65. +/- 3.37 L:4.54. +/-8.39. T:3.90. +/-5.69. AF: 1.18 1.76. − +/-5.30. − − +/-1.73. − +/-2.6RPUF P OL -BNC 0.3 wt.% 4.39. +/- 2.0 4.90. +/- 1.75 L:3.19. +/-1.34. T:2.55. +/-7.21. AF: 1.33 3.49. − +/-8.20. − − +/-1.52. − +/-2.8RPUF ISO - 4.05. +/- 3.36 L:5.11. +/-1.10. T:3.77. +/-9.30. AF: 1.29 - - 3.48. +/-3.2RPUF ISO -BNC 0.1 wt.% 4.15. +/- 3.9 4.01. +/- 2.29 L:4.77. +/-8.93. T:3.77. +/-7.83. AF: 1.21 4.04. − +/-8.66. − − +/-1.52. − +/-2.4RPUF ISO -BNC 0.3 wt.% - 4.34. +/- 3.52 L:3.08. +/-1.09. T:2.39. +/-3.73. AF: 1.29 2.57. − +/-8.92. − − +/-1.95. − +/-4.2RPUF ISO -BNC 0.5 wt.% 4.29. +/- 4.3 3.71. +/- 1.87 L:3.33. +/-7.22. T:2.00. +/-4.17. AF: 1.67 1.82. − +/-3.77. − − +/-1.42. − +/-4.612anostructured rigid polyurethane foams with improved specific thermo-mechanical properties using BacterialNanocellulose as a Hard Segment A P
REPRINT
Figure 1: Scheme of the experimental procedure used for the preparation of BNC reinforced RPUFs using the ISO andPOL routes. 13anostructured rigid polyurethane foams with improved specific thermo-mechanical properties using BacterialNanocellulose as a Hard Segment
A P
REPRINT
Figure 2: In-situ temperature rise as a function of processing route and BNC content.14anostructured rigid polyurethane foams with improved specific thermo-mechanical properties using BacterialNanocellulose as a Hard Segment
A P
REPRINT
Figure 3: NCO number as a function of BNC content.15anostructured rigid polyurethane foams with improved specific thermo-mechanical properties using BacterialNanocellulose as a Hard Segment
A P
REPRINT
Figure 4: ATR-FTIR absorption spectra of the RPUFs as a function of processing route and BNC content.16anostructured rigid polyurethane foams with improved specific thermo-mechanical properties using BacterialNanocellulose as a Hard Segment
A P
REPRINT
Figure 5: Heat flow as a function of temperature for the RPUF showing the first thermal cycle (down) and the secondthermal cycle (up). 17anostructured rigid polyurethane foams with improved specific thermo-mechanical properties using BacterialNanocellulose as a Hard Segment
A P
REPRINT
Figure 6: A) Compressive true stress as a function of deformation for the RPUF prepared with the POL route and as afunction of BNC concentration. B) Compressive true stress as a function of deformation for the RPUF prepared withthe ISO route and as a function of BNC concentration. Specific elastic modulus (C) and compressive strength (D) as afunction of processing route and BNC concentration. 18anostructured rigid polyurethane foams with improved specific thermo-mechanical properties using BacterialNanocellulose as a Hard Segment
A P
REPRINT
Figure 7: Specific elastic modulus (E’ sp ) and damping factor (tan /delta/delta