http://jrp.sagepub.com/Journal of Reinforced Plastics and Composites

http://jrp.sagepub.com/content/29/24/3543The online version of this article can be found at:

 DOI: 10.1177/0731684410381152

2010 29: 3543 originally published online 24 August 2010Journal of Reinforced Plastics and CompositesF. Cardona, H. Ku and L. Chouzenoux

preliminary resultsEffect of tannin on flexural properties of phenol formaldehyde glycerol reinforced composites:

  

Published by:

http://www.sagepublications.com

can be found at:Journal of Reinforced Plastics and CompositesAdditional services and information for    

  http://jrp.sagepub.com/cgi/alertsEmail Alerts:

 

http://jrp.sagepub.com/subscriptionsSubscriptions:  

http://www.sagepub.com/journalsReprints.navReprints:  

http://www.sagepub.com/journalsPermissions.navPermissions:  

http://jrp.sagepub.com/content/29/24/3543.refs.htmlCitations:  

What is This? 

- Aug 24, 2010 OnlineFirst Version of Record 

- Dec 13, 2010Version of Record >>

by guest on October 11, 2013jrp.sagepub.comDownloaded from by guest on October 11, 2013jrp.sagepub.comDownloaded from by guest on October 11, 2013jrp.sagepub.comDownloaded from by guest on October 11, 2013jrp.sagepub.comDownloaded from by guest on October 11, 2013jrp.sagepub.comDownloaded from by guest on October 11, 2013jrp.sagepub.comDownloaded from by guest on October 11, 2013jrp.sagepub.comDownloaded from by guest on October 11, 2013jrp.sagepub.comDownloaded from by guest on October 11, 2013jrp.sagepub.comDownloaded from by guest on October 11, 2013jrp.sagepub.comDownloaded from by guest on October 11, 2013jrp.sagepub.comDownloaded from by guest on October 11, 2013jrp.sagepub.comDownloaded from

Article

Effect of tannin on flexural propertiesof phenol formaldehyde glycerolreinforced composites:preliminary results

F Cardona1, H Ku1,2 and L Chouzenoux1

Abstract

Phenol formaldehyde composites are modified with glycerol–tannin mixes with varying percentages by weight of phenolic

resin to glycerol–tannin. The glycerol-to-tannin ratio is also varied. In this study also, different percentages by weight of

SLG (a commercial ceramic microsphere) are added to the composites. The effect of glycerol–tannin mix and added SLG

on the flexural properties was studied. It was found that the flexural stress decreases gradually with increasing tannin

when there is no SLG. With SLG, the presence of tannin in the composites reduces the flexural stress drastically. Tannin

reduces the strain at break. With SLG, this reduction of the yield strain is more apparent. Tannin increases the flexural

modulus of the composite and SLG acts in the opposite direction. The use of tannin in the phenolic composites improves

the ‘green’ character of the composite materials. It was also found that the presence of tannin made the mixing of

phenolic resins with glycerol easier.

Keywords

Phenolic resin, phenol formaldehyde, tannin, glycerol, SLG, flexural strength, strain at break, flexural modulus

Introduction

A research center in University of Southern Queensland(USQ) has begun work in the recent years to developnovel thermosetting resin toughening agents basedon bio-based sources. These additives have showngreat promise to date, being simultaneously inexpen-sive, highly effective, and environmentally beneficial.Bio-based tannins are the most preferred materialsfor toughening the phenolic resins, containing natu-ral phenol chemical functionalities, which have beenidentified to be of particular interest for their ‘green’character, readily availability in the market, andlow cost.

In this research project, the study was based on thedevelopment of phenolic and biodegradable additivecomposites with tannin obtained from different trees.Preliminary research work has shown that these tanninadditives can be successfully incorporated into com-mercial resole phenolic resins. However, many unan-swered technical questions remain regarding theaddition of these additives, especially in terms of the

mechanical performance of civil engineering compositesmanufactured using these biodegradable types of plas-ticizer additives. This study consisted of evaluating theflexural properties and the influence of different propor-tions of tannin mixed with glycerol and SLG in pheno-lic composites.

Samples are made with a view to understand thebehavior and the influence of different proportions oftannin and SLG in bio-based tannin phenolic compos-ites. Each sample is based on the weight of 100 g thatcorresponds to the mass of phenolic resin plus the mass

1Centre of Excellence in Engineered Fibre Composites (CEEFC),

University of Southern Queensland, Australia.2Faculty of Engineering and Surveying, University of Southern

Queensland, Australia.

Corresponding author:

Harry Ku, Faculty of Engineering and Surveying, University of Southern

Queensland, West Street, Toowoomba, 4350, Australia

Email: ku@usq.edu.au

Journal of Reinforced Plastics

and Composites

29(24) 3543–3553

! The Author(s) 2010

Reprints and permissions:

sagepub.co.uk/journalsPermissions.nav

DOI: 10.1177/0731684410381152

jrp.sagepub.com

of a mixture composed of glycerol (plasticizer) andtannin. After that, some SLG will be added. Five dif-ferent types of mixture are made and they are: MixtureA (100% glycerol and 0% tannin); Mixture B (90%glycerol and 10% tannin); Mixture C (80% glyceroland 20% tannin); Mixture D (70% glycerol and 30%tannin), and Mixture E (60% glycerol and 40%tannin). When SLG was used, it was necessary to adda thixotropic agent (powder Aerosil) to allow theproper dispersal of the SLG in the phenolic matrix.In 100 g of the resin and the glycerol–tannin mix(Mixtures A–E), samples with 0, 20, and 30 g of SLGwere prepared.

Three-point bending tests are carried out to evaluatethe flexural properties (flexural stress, strain at break,and flexural modulus) of the material. The samplesmade with Mixture C are tabulated as in Table 1.Phencat 15 is the catalyst used for the reaction. It isjust necessary to change the mixture and keep the sameproportions of the other constituents to obtain the 80samples.

Materials

Phenol formaldehyde and its catalyst

The commercial resole resin used in this study wasJ-2027L, produced by Borden Ltd. Its official name isBorden Cellobond J2027L.1 The acid catalyst used tocross-link the resin was Phencat 15, which is a fastaction acid catalyst. Its composition is shown in

Table 2. There is a strong exothermic reaction whenphenolic resin is mixed with catalyst. The ratio byweight of the resin to hardener for all samples in thisstudy was chosen to be 50 : 1. The phenolic BordenCellobond J2027L is a brown viscous liquid with acharacteristic phenolic smell. While using the resin, itis necessary to wear a mask as the gas emissions fromthe resin are noxious. This resin is more commonlycalled resole phenolic resin.2

Table 3 shows the composition of Borden CellobondJ2027L. It is composed from the reaction of phenol andformaldehyde; these two chemical products react toform a hydroxymethyl derivative phenol. But the reac-tion is incomplete and therefore unreacted formalde-hyde and phenol, and also water that evolves duringthe condensation reaction between phenol and formal-dehyde remain.

With reference to phenolic molecule of Figure 1,there are five hydrogen atoms in the benzene ring butbecause of limited space, there are only three possiblesites for reaction and the phenolic molecule is said tohave a functionality of three.3,4 As the functionality ofthe phenolic molecules is greater than two, the mole-cules can react with formaldehyde molecules to form a3D network polymer.5

Biodiesel glycerol

Biodiesel glycerol is a plasticizer agent for phenolicresins and is obtained as a sub-product from the pro-duction of biodiesel using common vegetable oils. As aplasticizer agent when added to phenolic composites itincreases the elasticity and toughness of the phenolicmaterials. This is because the presence of glycerol inthe phenolic network reduces the characteristic brittle-ness of phenolic resins by decreasing the cross-link den-sity of the cured composites.

Table 1. Proportions of the 16 samples for Mix C

Phenolic

resin (g) Mix C (g) SLG (g) Phencat 15 (g) Aerosil (g)

90 10 – 5 –

80 20 – 7 –

70 30 – 8 –

60 40 – 10 –

90 10 20 7 2

80 20 20 9 2

70 30 20 10 2

60 40 20 12 2

90 10 30 7 1

80 20 30 9 1

70 30 30 10 1

60 40 30 12 1

90 10 40 7 1

80 20 40 9 1

70 30 40 10 1

60 40 40 12 1

Table 3. Composition of Borden Cellobond J2027 L

Name Percentage

Phenol/Formaldehyde resin 30–60

Phenol 1–10

Formaldehyde 1–5

Water 10–12

Table 2. Composition of Phencat 15

Name Percentage

Xylenesulfonic acid 70–90

Phosphoric acid 10–20

Water 1–10

3544 Journal of Reinforced Plastics and Composites 29(24)

The bio-based tannin (Bondtite 345)

It is wattle tannin and is supplied by Bondtite Co. Ltd.in Australia. This tannin is a dark brown powder,which contains moisture of less than 10%. Tannin isan excellent renewable resource for replacing petro-leum-derived phenolic compounds. The major woodspecies from which it can be obtained are mimosa,quebracho, and radiata pine. The structure of themain polymeric constituents of wattle tannin can berepresented in Figure 2. Because of their size andshape, the tannin molecules become immobile at alow level of condensation with formaldehyde andthe available reactive sites are too far apart for fur-ther methylene bridge formation. The result is incom-plete polymerization and therefore weakness.

HO OH

OH OH

H

H

H

H

H

H

H

H

HH H

H

H2O

H

H

H H

H H

H H

H H

O

CFormaldehyde

Phenols

Phenolformaldehyde

C

C

C C

C C

C

C

C C

C C

C

C

C

C

+

C

C C

C

C

C C

C C

Figure 1. Formation of phenol formaldehyde.

(OH)

(OH)

OH

OH

OH

HO O

A

B

7

6 105

89

1 23

4

1′2′

3′

4′

5′6′

Figure 2. Structure of wattle tannin.

Cardona et al. 3545

Bridging agents with longer molecules, such as phenol/formaldehyde condensates, have been shown to becapable of bridging the distances that are too largefor methylene bridges.6

E-spheres, SLG

Envirospheres SLG (E-spheres) is a commercialceramic microsphere product obtained as a fly ashby-product. The particle size of this general purposeE-spheres ranges from 20 to 300mm with approximatemean of 130mm. The relative density of E-spheres is 0.7.E-spheres are a combination of silica, SiO2 (55–60%),alumina, Al2O3 (36–44%), iron oxide, Fe2O3 (0.4–0.5%),and titanium dioxide, TiO2 (1.4–1.6%).

Aerosil COK 84

It is a white powder which is based on a mix of hydro-philic fumed silica and highly dispersed fumed alumin-ium oxide in the ratio of 5 : 1. Normally, Aerosil isapplied as a thixotropic agent in liquid systems toimprove their rheological properties. In this study, itfacilitates the homogeneous dispersal of SLG in thesamples.

Sample preparation

Mixture of glycerol and tannin

Five different mixes (A–E, as mentioned above) weremade in this study. For each mix, 700 g was prepared.In order to facilitate the mixture between the glyceroland the tannin, 1 g of sulfuric acid was added (exceptfor Mix A) for each 100 g of mixture. The proportionsof the mix are shown in Table 4. To prepare the mix,glycerol and sulfuric acid were mixed at a temperaturebetween 80�C and 100�C. After that, the tannin wascarefully added to the mixture of glycerol and sulfuricacid; the mixture was stirred until there was no moretannin left; stirring was continued for the next45–60min. Higher the content of tanning, higher wasthe viscosity of the mix.

The samples

Each sample has been prepared in a round plastic con-tainer (diameter: 9 cm and height: &8 cm) and all sam-ples have the same protocol of preparation; first, thephenolic resin was put into the container; then, theglycerol–tannin mixture was added followed by stirringthe two liquid components until the mix was homoge-neous (usually after 3–5min of stirring the mix).

After that, there were two cases of dealing withsamples:

. If there was no SLG, the catalyst was added to themixture of the two liquids and then they were mixedquickly for 3 or 4min.

. If there was SLG, the E-spheres and the Aerosil wereadded to the mixture of the two liquids and thenthey were mixed until a homogeneous mixture wasobserved (3–5min); after this, the catalyst was addedand further mixed quickly for additional 3–4min.

The quantity of Aerosil added depended on the SLGadded. With more SLG, for example, 40 g, moreAerosil, e.g. 2 g, had to be added in order to have agood dispersal of the SLG in the sample. For 20 and30 g of SLG, 1 g of Aerosil was enough. The sampleswere initially left to cure at room temperature over-night. More the tannin in the sample, more the catalystto be added.

Post-curing the samples

After curing the samples overnight at ambient condi-tions, the samples were post-cured in an oven for 4 h at80�C. Any traces of water left behind would be drivenout by this curing process. It is important to addenough quantity of catalyst to the samples to hardenthem before putting in the oven; if insufficient quantityof catalyst is added, the samples would still be soft evenafter they were cured in ambient conditions for onenight. On the other hand, if excessive catalyst isadded, the curing would be fast, highly exothermicand bubbles would evolve, resulting in an unusablematerial.

Samples and testing

When a sample is taken out of the plastic container, thesample has a round shape and its height depends on thequantity of SLG. The height ranges from 1 cm (withoutSLG) to 2.5 cm (with 40 g of SLG). In order to obtain asample which can be used for flexural strength analysis,it is necessary to cut it to the required dimensions.

With the circular saw, four or five test samples werecut to the same thickness. The standard used for the

Table 4. Composition in mass (g) of each mix

Glycerol Tannin Sulfuric acid

Mix A (100-0) 700 0 0

Mix B (90-10) 630 70 7

Mix C (80-20) 560 140 7

Mix D (70-30) 490 210 7

Mix E (60-40) 420 280 7

3546 Journal of Reinforced Plastics and Composites 29(24)

mechanical test is ISO 14125:1998(E), which allows theresults to be compared with the published literature ofsimilar polymeric materials.7 A Material TestingSystems (MTS) 810 was used for the tests and the soft-ware used was TESTWORKS 4. The dimensions of thespecimens of resins were 100� 10� 4mm3 and tested ata cross-head speed of 1mm/min.

Results and discussion

Three-point bending tests were used to evaluate theflexural properties of the composites.8 As mentioned,the tests were carried out in accordance with ISO14125:1998(E). The results will be presented in threedifferent parts, flexural stress, strain at break, and flex-ural modulus.

Effect of glycerol–tannin on flexural strength

Figure 3 shows the flexural strength of phenolic com-posites without SLG. The values of the flexural strengthof composites with no tannin were higher than theircounterparts. However, the difference in the values ofthe flexural strength of composites with Mixtures A(0% of tannin) and B (with 10% of tannin) was mini-mal. Those values of composites with Mixture B werewithin the 5% markers of the values of composite withMixture A. However, the values the flexural stress ofthe composites with Mixture E (40% tannin) were atleast 24% lower than those with Mixture A; this indi-cates that the glycerol plays a significant role as a plas-ticizer agent inside the phenolic network, while the

increasing the addition of tannin mixed with glycerolclearly has the opposite effect to glycerol on the flexuralstrength of the phenolic network.

The results indicate that the presence of tannindecreased the flexural strength of the composites. Thepeak flexural strength of composites with Mixture Awas 49MPa when the percentage by weight of the mix-ture was 21.5%. The peak flexural strength of compos-ites with Mixture B was 47MPa when the percentage byweight of the mixture was 26%. The peak flexuralstrength of composites with Mixture E was 26MPawhen the percentage by weight of the mixture was31.5%. Clearly, as the amount of tannin mixed withglycerol increased and also the amount of the glyc-erol–tannin in the phenolic resin increased, there is acontinuous decrease in the flexural strength of thecomposites.

Figure 4 shows the flexural strength of phenoliccomposites with 20% by weight of SLG. It is evidentthat the values of flexural strength are much lower thantheir counterparts in Figure 3. This is totally due to thepresence of SLG in the mixtures as previouslyreported.9 In addition, the values also dropped linearlywith increasing percentage by weight of mixtures in thesamples; this applies to all mixtures (A–E). The decreas-ing trend was even larger with composites with MixtureA (no tannin). Up to 30% by weight of mixtures, thevalues of flexural strength of the materials withMixtures B and E were within 5% markers of thosewith Mixture A as depicted in Figure 4; clearly, in thecomposites with SLG filler the increasing the additionof tannin to the glycerol–tannin mix appears to

0

10

20

30

40

50

60

0

Percentage by weight of mixtures in the samples

Fle

xura

l str

ess

(MP

a)

Mix A (100–Glycerol 0–Tannin)

Mix B (90–Glycerol 10–Tannin)

Mix E (60–Glycerol 40–Tannin)

45403530252015105

Figure 3. Flexural stress of phenolic composite with 0% SLG.

Cardona et al. 3547

improves the flexural strength of the composites. Thiseffect may be due to an improvement in the interfaceinteraction between the SLG filler and the phenolicmatrix by the added tannin. However, further work isrequired to fully clarify this effect of tannin in thecomposites.

It should be noted that the addition of SLG reducedthe flexural strength of the composite drastically; thepeak flexural stress of composite with Mixture B and0% SLG was 45.8MPa while the strength of the equiv-alent sample with 20% of SLG was 9.1MPa; the dropcorresponds to about 80% of the initial value. Thisindicates that SLG acts a filler inside the resin matrix,not as a reinforcing agent, and therefore the purpose ofadding SLG to the composites is to reduce the overallcost (SLG is quite cheap in comparison with the cost ofthe resin), but there is a proportional decrease in themechanical performance of the materials with increas-ing SLG content. Also, the results indicate that thisnegative effect of SLG on the mechanical propertiesof the composites can be partially compensated by theaddition of tannin dissolved in glycerol to thecomposites.

Figure 5 shows the flexural strength of phenoliccomposites with glycerol–tannin with 30% SLG; itwas observed that the flexural strength decreasedslightly with increasing mixture by weight; the decreas-ing trend was more significant for the composites withMixture A (no tannin), similarly to the samples with20% SLG content, as shown in Figure 4. The increasein SLG further reduced the flexural strength of thecomposites; the peak flexural strength of compositeswith 30% of SLG and Mixture E was 10.5MPa,

while that with 20% of SLG and the same mixturewas 15.08MPa; this is a drop of 30%. The increase oftannin content in the composites samples with 30%SLG content resulted in a proportional increase ofthe flexural strength as shown in Figure 5; this confirmsthat the increasing presence of tannin in the network ofthe composites with SLG increases the flexural strengthof the materials.

Effect of glycerol–tannin on strain at break (%)

Figure 6 shows the strain at break (%) of phenoliccomposite reinforced with glycerol–tannin with 0%(without) SLG; it can be found that the strain atbreak increased with the amount of glycerol–tanninadded to the phenolic resin and peaked at some pointbefore dropping back when the tannin by weight wasover 30%. The composite with Mixture A was used as areference and its strain at break peaked at 3.8% whenMixture A was 31.5% by weight; the results confirmthat the increasing addition of glycerol resulted in aproportional increase of the strain at break up to30% addition to the resin, but above this value thereis a decreases in the strain at break values with furtheraddition of glycerol. Therefore, for composites withoutSLG, the presence of glycerol–tannin in the phenolicnetwork increased the flexural strength and the strainat break, which as previously mentioned is due to thetoughening effect of the glycerol–tannin inside thecured phenolic network. With added SLG filler(Figures 7 and 8), there is an initial increase in thevalues of the strain at break with increasing glyceroland glycerol–tannin mix up to 20%. However, further

0

4

8

12

16

20

Percent by weight of mixtures in the samples

Fle

xura

l str

ess

(MP

a)

Mix A (100–Glycerol 0–Tannin)

Mix B (90–Glycerol 10–Tannin)

Mix E (60–Glycerol 40–Tannin)

0 45403530252015105

Figure 4. Flexural stress of phenolic composite with 20% SLG.

3548 Journal of Reinforced Plastics and Composites 29(24)

increase in the amounts of glycerol–tannin decreasesthe strain at break of the composites; with Mixture E(40% tannin) the strain at break was at least 66% lowerthan those with Mixture A. This effect was quite similarto that of the glycerol–tannin mix on the flexural stressvalues of the composites. Therefore, the results clearlyindicate that in the absence of SLG filler, the increasingamount of tannin in the mix with glycerol in the com-posites reduces the flexural strength and flexural strainof the materials. However, with added SLG filler theeffect of tannin in the mechanical properties is more

complicated; the strain at break initially increasedwith the amount of tannin, but the strain at breakdecreased with increasing glycerol–tannin when theamount of the mix in the composites was above the20% mark, as shown in Figures 7 and 8. It shouldalso be mentioned that the addition of SLG fillerreduced the strain at break in comparison with the sam-ples without SLG, similar to the above reported resultsof the flexural strength of the composites.

Figure 8 shows the strain at break (%) of phenoliccomposite reinforced with glycerol–tannin with 30%

0

2

4

6

8

10

12

Percentage by weight of mixtures in the samples

Fle

xura

l str

ess

(MP

a)

Mix A (100–Glycerol 0–Tannin)

Mix B (90–Glycerol 10–Tannin)

Mix E (60–Glycerol 40–Tannin)

0 45403530252015105

Figure 5. Flexural stress of phenolic composite with 30% SLG.

0

1

2

3

4

Percentage by weight of mixtures in the samples

Str

ain

at b

reak

(%

)

Mix A (100–Glycerol 0–Tannin)

Mix B (90–Glycerol 10–Tannin)

Mix E (60–Glycerol 40–Tannin)

0 45403530252015105

Figure 6. Strain at break (%) of phenolic composite with 0% SLG.

Cardona et al. 3549

SLG; similar to the results presented in Figure 7 for thesamples with 20% SLG, it was found that the strain atbreak increased slightly with increasing percentage byweight of mixture and peaked at some point beforedropping back when the mixture by weight was over25%. This confirms that the increases in SLG fillerresult in a proportional decrease in the strain at breakof the composites. The increasing amount of tannin inthe mix with glycerol resulted in even lower values ofthe strain at break as, depicted in Figures 7 and 8.

Effect of glycerol–tannin on flexural modulus

Figure 9 shows the flexural modulus of phenolic com-posite reinforced with glycerol–tannin with 0% SLG;it can be found that the flexural modulus decreasedwith increasing percentage by weight of the mixture.Without SLG, the flexural modulus at 40% by weightof Mixture A was only 1.05GPa; with 30% by weightof SLG, its value was 1.78GPa, an increase of 70%.The lowest flexural modulus was found in composite

0

0.5

1

1.5

2

Percentage by weight of mixtures in the samples

Str

ain

at b

reak

(%

)

Mix A (100–Glycerol 0–Tannin)

Mix B (90–Glycerol 10–Tannin)

Mix E (60–Glycerol 40–Tannin)

0 45403530252015105

Figure 7. Strain at break (%) of phenolic composite with 20% SLG.

0

0.5

1

1.5

2

Percentage by weight of mixtures in the samples

Str

ain

at b

reak

(%

)

Mix A (100–Glycerol 0–Tannin)

Mix B (90–Glycerol 10–Tannin)

Mix E (60–Glycerol 40–Tannin)

0 45403530252015105

Figure 8. Strain at break (%) of phenolic composite with 30% SLG.

3550 Journal of Reinforced Plastics and Composites 29(24)

with Mixture A and percentage by weight of mixture of40%; its value was 1.025GPa. Therefore, it is evidentthat with an increase in the glycerol content in the com-posites there is a proportional decrease of the flexuralmodulus, but with an increase in the amount of tannindissolved in glycerol in the composite formulation thereis an increases of the flexural modulus in comparisonwith the composites without tannin. This can beexplained in terms of the effect of tannin on the inter-face between the SLG filler and the phenolic resin,

where the tannin appears to improve this interactionbetween the filler and the matrix.10

Figure 10 shows the flexural modulus of phenoliccomposites reinforced with glycerol–tannin and with20% SLG; it was found that the flexural modulusdecreased with increasing percentage by weight of themixture. With 20% SLG, the values of the flexuralmodulus continuously decreased irrespective of the per-centage by weight of the mixture. The lowest value ofthe flexural modulus was found in the composites with

0

0.5

1

1.5

2

2.5

3

Percentage by weight of mixtures in the samples

Fle

xura

l mod

ulus

(G

Pa)

Mix A (100–Glycerol 0–Tannin)

Mix B (90–Glycerol 10–Tannin)

Mix E (60–Glycerol 40–Tannin)

0 45403530252015105

Figure 9. Flexural modulus of phenolic composite with 0% SLG.

0

0.5

1

1.5

2

2.5

Percentage by weight of mixtures in the samples

Fle

xura

l mod

ulus

(G

Pa)

Mix A (100–Glycerol 0–Tannin)

Mix B (90–Glycerol 10–Tannin)

Mix E (60–Glycerol 40–Tannin)

0 45403530252015105

Figure 10. Flexural modulus of phenolic composite with 20% SLG.

Cardona et al. 3551

Mixture A and percentage by weight of the tannin/glyc-erol mixture of 40%; however, the increase of tannin inthe composites formulation resulted in a proportionalincrease in the flexural modulus, similar to the resultswithout SLG, as shown in Figure 9.

Figure 11 shows the flexural modulus of phenoliccomposites reinforced with glycerol–tannin and with30% SLG; it can be observed that the flexural modulusdecreased with increasing percentage by weight of themixture. With 30% SLG, the values of the flexuralmodulus were lower than the values of the compositeswith only 20% of SLG (Figure 10) or without SLG(Figure 9). Similar to the previous results, the increasein tannin dissolved in glycerol resulted in a propor-tional increase in the flexural modulus of the compos-ites. Considering the overall mechanical performance ofthe composites with only glycerol–tannin and withoutSLG filler (Figures 3, 6, and 9), there is a decrease in theflexural strength and flexural strain with increasingtannin content, but the results showed the oppositebehavior for the flexural modulus where there is anincrease with the amount of tannin dissolved. This isnot surprising as the values of flexural modulus areobtained by dividing the flexural stress by the flexuralstrain. It can be argued that even the trend includingflexural modulus was consistent and the values of theflexural stress, strain, and modulus are therefore reli-able. With the presence of 20% of added SLG (Figures4, 7, and 10), the trend is quite similar to the sampleswithout the filler, except for the flexural strength thatclearly decreased as the amount of glycerol–tannin mix-ture increased. For the values of the samples withhigher SLG content (30%; Figures 5, 8, and 11), the

behavior of the mechanical properties is more compli-cated with the flexural modulus decreasing as theglycerol–tannin mixture is increased, but a high vari-ability in their values of the flexural strength and flex-ural strain was observed. However, there is a definedincrease in the flexural strength as the amount of dis-solved tannin in the composites increased, which isaccompanied by a decrease in the flexural strain forthe same samples. Moreover, tannin can also partiallyreplace the phenol in phenol formaldehyde resin asreported by other researchers.11–13 This is critical aspetroleum crisis is surging again presently.

Conclusions

Without SLG filler in the composites, the flexuralstrength reaches a maximum of 20–25% of addedglycerol–tannin mixture, but further increase is fol-lowed by a decrease in the flexural strength of the com-posites. Also, there is a decrease in the flexural strengthwith increasing amount of tannin in the formulations.With SLG filler in the composites, the increasingamount of glycerol resulted in a decrease of the flexuralstrength, but this effect was compensated as the amountof dissolved tannin increased in the compositeformulations.

The increase of tannin in the composites was alsoaccompanied with a decrease in the flexural strain atbreak, with and without SLG, which indicates a molec-ular interaction (cross-linking) between the tannin andthe phenolic resin. As a result of this molecular inter-action between tannin and phenolic resin, the flexuralmodulus increased with higher tannin content in the

0

0.2

0.4

0.6

0.8

1

1.2

Percentage by weight of mixtures in the samples

Fle

xura

l mod

ulus

(G

Pa)

Mix A (100–Glycerol 0–Tannin)

Mix B (90–Glycerol 10–Tannin)

Mix E (60–Glycerol 40–Tannin)

0 45403530252015105

Figure 11. Flexural modulus of phenolic composite with 30% SLG.

3552 Journal of Reinforced Plastics and Composites 29(24)

composites. Finally, the use of tannin (a renewable anda biodegradable additive) as a toughening agent in phe-nolic composites increases the ‘green’ character of thematerials, making them more attractive as a structuralmaterial in future civil engineering applications. This isparticularly true when tannin is used to partiallyreplace the phenol in phenol formaldehyde resin.11,12

Funding

This research received no specific grant from any fund-ing agency in the public, commercial, or not-for-profitsectors.

References

1. Chemwatch. Material safety data sheet for Borden

Cellobond J2027L, 2005.2. Chemwatch. Material safety data sheet for Borden

Phencat 15, 2005.

3. Strong AB. Plastics: materials and processing, 3rd ed.New York: Pearson/Prentice-Hall, 2006.

4. WangH.Engineeringmaterials – study book 1. Toowoomba,

Queensland: University of Southern Queensland, 2008.5. Shackelford JF. Introduction to materials science for engi-

neers. New York: Macmillan, 1992.

6. Ge J, Zhong W, Guo Z, Li W and Sakai K.

Biodegradable polyurethane materials from bark and

starch. I. Highly resilient foams. J Appl Polym Sci 2000;

77: 2575–2580.7. ISO, ISO 14125:1998(E). Fibre reinforced plastic compos-

ites – determination of flexural properties. Geneve, Austria:

International Organization for Standardization, 1998.

8. Callister WD. Materials science and engineering: an intro-

duction. New York: Wiley & Sons, Inc., 2007.

9. Cardona F, Chouzenoux L and Ku H. Effect of tannin on

thermal properties of phenol formaldehyde glycerol rein-

forced composites: preliminary results. J Appl Polym Sci

2010; (submitted for publication).10. Ku H, Rogers D, Davey R, Cardona F and Trada M.

Fracture toughness of phenol formaldehyde composites:

pilot study. J Mater Eng Perform 2008; 17(1): 85–90.11. Multu I, Alma MH, Basturk MA and Oner C.

Preparation and characterization of brake linings from

modified tannin–phenol formaldehyde resin and asbes-

tos-free fillers. J Mater Sci 2005; 40: 3003–3005.12. Nimz H. Wood adhesives, chemistry and technology.

New York: Marcel Dekker Inc., 1983.13. Ono HK and Inoue A. Adhesives and bonded wood prod-

ucts. Madison: Forest Products Society, 1994.

Cardona et al. 3553