ilable at ScienceDirect

Polymer Testing 28 (2009) 296–300

Contents lists ava

Polymer Testing

journal homepage: www.elsevier .com/locate/polytest

Material Behaviour

Effect of UV radiation on some polymeric networksbased on vinyl ester resin and modified lignin

Liliana Rosu, Constantin N. Cascaval*, Dan Rosu‘‘Petru Poni’’ Institute of Macromolecular Chemistry, Iasi, Romania

a r t i c l e i n f o

Article history:Received 11 November 2008Accepted 5 January 2009

Keywords:Vinyl ester resinLigninSemi-IPNsPhotochemical stability

* Corresponding author. Tel.: þ40 232 217 454; faE-mail address: cascaval@icmpp.ro (C.N. Cascava

0142-9418/$ – see front matter � 2009 Elsevier Ltddoi:10.1016/j.polymertesting.2009.01.004

a b s t r a c t

Some semi-interpenetrating polymer networks (semi-IPNs) based on vinyl ester resin(VER) and ammonium lignosulfonate (ALS) modified lignin were synthesized and char-acterized using Fourier transform-infrared (FT-IR) spectroscopy, optical microscopy (OM)and differential scanning calorimetry (DSC) techniques. VER was synthesized starting froman epoxy resin in reaction with acrylic acid. The cross-linking reaction was initiated by UVradiation. The synthesized networks showed good compatibility, due to some possibleinteractions between the functional groups from VER and ALS components (OH, espe-cially). A slight effect of photostabilization of the VER was noticed, due to the ALS struc-tures which were incorporated into the resin matrix.

� 2009 Elsevier Ltd. All rights reserved.

1. Introduction

Vinyl ester resins (VERs), very well known thermor-eactive polymers, are obtained from the reaction betweena,b-unsaturated acids and epoxy resins. They are charac-terized by the presence of the double bonds as the endgroups, which are derived from the acid structure. Theexistence of the double bonds in VER structures is a way toobtain cross-linked networks by initiation with free radi-cals. Due to their excellent mechanical properties, thermalstability and high corrosion resistance, VERs are the poly-mer matrices used for manufacturing high performancecomposites intended for marine, aerospace, transportation,building, construction and biomedical applications [1–6].

VERs and their compositions are widely used asprotective films for various materials. Unfortunately, theiruse as surface coatings in outdoor applications is limitedby the sensitivity to photo-oxidative degradation [7].Blending of VERs with other synthetic and/or naturalpolymers provides a means of producing new materialswhich combine the useful properties of all the constituents

x: þ40 232 211 299.l).

. All rights reserved.

[8,9]. Interpenetrating polymer networks (IPNs) are ideallycompositions of two (or more) chemically distinct poly-meric networks held together exclusively by their perma-nent mutual entanglements [10]. This definition has beengeneralized to include semi-IPNs, when only one of thecomponents forms a network [11].

Lignin is an abundant amorphous natural polymer,readily available and relatively inexpensive. Because of itsstructure and macromolecular properties, lignin is aninteresting material for use as a component in polymerblends [12], being used as stabilizer (antioxidant) forplastics and rubber, as well as in the formulation ofdispersants, adhesives and surfactants [13,14]. Due to itsphenolic structures, lignin is an excellent light absorber.

Ammonium lignosulfonate (ALS), a soluble derivate oflignin, is a by-product of the acid sulfate pulp-makingprocess. Like lignin, ALS has very complex macromolecularstructure, with negatively charged sulphonate, hydroxyl,phenolic and carbonyl groups.

Also, VERs are considered as corrosion resistant mate-rials, but their use in outdoors is strongly limited bysensitivity to visible and ultraviolet domains. Lignin and itsderivative (ALS) can be used as antioxidant agents in thecompositions with VER resin. In the presence of UV radia-tion, VERs undergo severe degradation with changes in the

OH

- O - CH2 - CH2CH

CH3

CH3

- C -- O-CH2-CH-CH2-O-

CH3

CH3

- C --CH2 -O - -CHCH2

CH = CH2 CH2 = CH

CO

O

CO

O

OH OHn

n

O CH2OH2C - C -

CH3

CH3

- O-CH2-CH-CH2-O- - C -

CH3

CH3OH

CH2HC

O

H2C CH

O

+

CH2 CH COOH2

BTAC

Scheme 1. Synthesis of VER.

Table 2IR characteristic bands of VER–ALS-3 semi-IPN sample.

Wavenumber (cm�1) Main assignment

3320 nOH associated3036 nCH of the aromatic ring2960/2929/2871 nCH; nCH2; nCH3

1889 gCH aromatic ring1712 nC]O1635 nC]C double bond

L. Rosu et al. / Polymer Testing 28 (2009) 296–300 297

surface chemistry, morphology and structure [15]. The aimof this study was to investigate the behaviour of somesemi-IPNs based on VER and ALS to UV radiation.

2. Experimental part

2.1. VER synthesis

The chemical reactions used for synthesis of VER areshown in Scheme 1.

The VER based on bisphenol A was obtained by theprocedure described elsewhere [16]. The synthesis wascarried out using commercial Ropoxid 501 resin (PolicolorSA Bucharest, Romania) and acrylic acid, in the presence ofbenzyltributylammonium chloride (BTAC) as catalyst.Ropoxid 501 resin, with epoxy equivalent value0.525 equiv. 100 g�1 and number-average molecular weightðMnÞ of 381 g mol�1, was obtained by reaction of bisphenolA with epichlorohydrin (EPI). The synthesized VER, solublein dimethylformamide (DMF), is characterized byMn ¼ 525 g mol�1 and melting point 53 �C.

2.2. ALS characterization

ALS is a by-product of the acid sulfate pulp makingprocess, with characteristics as follows: amount of solidsubstances, between 45 and 48 wt%; OCH3 groups,8.5–9.2 wt%; COOH groups, 6–18 wt%; elemental analysis:C, 45.05%; H, 5.5%; N, 3.6%; S, 6.5%. UV-VIS spectra recordedfor ALS water solutions evidence the presence of three

Table 1VER and VER–ALS semi-IPNs.

Sample VER content (%) ALS content (%)

VER 100 –VER–ALS-1 99 1VER–ALS-2 97 3VER–ALS-3 95 5VER–ALS-4 93 7

absorption maxima, which are positioned at wavelengthslower than 300 nm (l ¼ 201.7 nm, A ¼ 1.4472; l ¼ 262.85,A ¼ 0.1746; l ¼ 280.85, A ¼ 0.2159).

2.3. VER–ALS semi-IPNs preparation

The VER–ALS semi-IPNs were prepared by a sequentialprocedure of mixing VER and ALS solutions in DMF.Bis(2,4,6-trimethylbenzoyl)phosphine oxide, in concentra-tion of 0.25 wt% against the VER mass, was added to themixture as photoinitiator. The VER–ALS composites wereobtained by casting the mixtures onto glass slides and,subsequently, drying in air and then in vacuum at 110 �C for3 h. The solvent was completely removed and the filmswere cross-linked in the presence of UV radiation.

2.4. Irradiation and analysis

The UV irradiation of the obtained VER–ALS semi-IPNs was carried out by means of a medium-pressure

1591 nC]C aromatic ring1488 dCH2 or/and dCH31467 Aromatic ring stretch1382 dCH2 or/and dCH of the double bond1293 nC–O1224 nC–O–C1160 nC–CO–O1090 dCH aromatic ring940 dCH double bond812 Polyhydroxyether backbone750 gCH aromatic ring

ALS

CH2

OH

- O - - CH- C -- O- -CH- -O-- C -- -O - -CHCH2

CH2 CH2 CH2 CH2

CH3

CH3

CH3

CH3CH3

CH3

CH3

CH3

CH2 CH2

CH2

CH2 CH2

CH2

CH2

CH2CH2CH2

OH OHn

O CO CHCO O

CH2

HC

OH

- O - - CH

CH3

- C -- O- -CH- -O-- C -- -O - -CH

OH OHn

O CO CHCO O

CH2

HC

ALS

ALS

ALS ALS

ALS ALS

ALS

ALS

Fig. 1. VER–ALS semi-IPNs formulation.

L. Rosu et al. / Polymer Testing 28 (2009) 296–300298

mercury lamp, HQE-40 type, having a polychromeemission spectrum in the range between 240 and570 nm. The more energetic radiation (not present innatural light) was eliminated using a 30 mm borosilicateglass filter. The samples, as films, were mounted ona rotating device, which was positioned at a distance of60 mm from the lamp. The temperature inside theirradiation chamber was kept at 40–45 �C by means ofa fan and a distilled water filter. The mounted filmswere withdrawn from the device at different times andanalyzed using FT-IR spectroscopy, DSC and OMtechniques.

The FT-IR spectra were recorded before and afterirradiation using a FT-IR spectrophotometer, BrukerVertex 70 type, at a nominal resolution of 4 cm�1. TheOM micrographs were obtained by means of a MC5A(�300) optical microscope (IOR-Bucharest, Romania) atroom temperature. The thermal measurements werecarried out with a 12E Mettler calorimeter operated inscanning mode (10 �C min�1), nonpurified nitrogenatmosphere and temperature range between 20 and40 �C.

Fig. 2. DSC thermograms of VER–ALS semi-IPNs:VER (C), VER–ALS-1 (B),VER–ALS-2 (-), VER–ALS-3 (,), VER–ALS-4 (:).

3. Results and discussion

The VER–ALS composites with different concentrationsof ALS produced as described above are listed in Table 1.

The FT-IR spectra of the obtained semi-IPNs were verysimilar to those of the crude VER [7]. Table 2 summarizesthe most representative IR bands for one of the semi-IPNsamples (VER–ALS-3).

As a result of the cross-linking reaction, the vibrations,which are characteristic of the double bond (1635 and940 cm�1), disappeared in VER–ALS semi-IPNs. Simulta-neously, the band at 1712 cm�1, specific to the carbonylstretching vibration in ester group (nC]0), is shifted to1727 cm�1, the latter being characteristic of the saturatedester structures. The disappearance of the double bondsignals and displacement of the carbonyl absorbance areproofs for the change of the VER structure in the presenceof ALS. An exemplification of the manner in which lignin, assuch, is inserted through the chains of the cross-linked VERnetwork is shown in Fig. 1.

The DSC technique was used to examine both thecompatibility of the components inside the studiednetworks and to evidence the possible intermolecularinteractions. Fig. 2 shows the DSC thermograms, which arecharacteristic of the synthesized interpenetrating polymernetworks.

The glass transition temperatures (Tgs) of the VER–ALSsemi-IPNs were evaluated by interpolation the middle ofthe transition interval, which was determined using thetangent method. The Tg values determined for the studiednetworks are listed in Table 3.

A short examination of the curves in Fig. 2 and data inTable 3 show that both the etalon sample and the analyzedVER–ALS semi-IPNs are characterized by a single Tg value,which is dependent on composition. This behaviour

Table 3Tg values of the synthesized samples.

Sample Tg (�C) Transition interval (�C)

VER 55 6VER–ALS-1 63 9VER–ALS-2 73 13VER–ALS-3 79 15VER–ALS-4 82 19

Fig. 3. OM micrographs of VER and of two VER–ALS semi-IPN samples.

L. Rosu et al. / Polymer Testing 28 (2009) 296–300 299

demonstrates good compatibility of the components insidethe studied networks [17,18]. Simultaneously, the enlarge-ment of the glass transition interval, together with increaseof ALS content in the networks, is an indication of goodinterpenetration of the components [19,20]. The morpho-logical modifications inside the obtained networks wereobserved by OM in transmitted light. The micrographsobtained for VER (etalon sample) and for VER–ALS semi-IPNs (VER–ALS-2 and VER–ALS-4) are shown in Fig. 3.

Fig. 3 shows that the etalon sample is presented asa continuous smooth surface, which is penetrated bya reduced number of craters and pinholes, which can beexplained by the presence of some imperfections registeredwhen the films were deposited on the plates. In compar-ison with the etalon sample, the VER–ALS semi-IPNs showa biphasic morphology, where the ALS phase (more darkzones) is uniformly dispersed in the VER matrix.

The FT-IR spectra of the synthesized networks recordedafter UV irradiation (l> 300 nm) led to important chemicalmodifications in the polymer structure. As an example,Fig. 4 shows the spectrum difference obtained by subtrac-tion of the absorbances from the spectrum recorded for UV20 h irradiated VER–ALS-4 semi-IPN sample from a non-irradiated sample.

The FT-IR spectrum shown in Fig. 4 has negative andpositive intensities. The negative absorbances reflect the

4000 3500 3000 2500 2000 1500 1000

Relative ab

so

rb

an

ce

Wavenumber (cm-1

)

-0.5

+0.5

0

3600 - 3100

30362871

2992

1727

1680

1748

1591

1625

1488

15251188

12411329 1141

1053 851

812

750

Fig. 4. IR spectrum difference for VER–ALS-4 semi-IPN sample.

structures that were formed during the UV irradiation,while the positive absorbances mark those structures thatwere lost. The positive absorbances at 3036, 2871, 1727,1591, 1488 and 821 cm�1 can be attributed to the aromaticstructures from the saturated ester backbone. At this timeof irradiation the synthesized VER–ALS semi-IPN samplesshow mainly negative peaks, corresponding to the newstructures that were formed during the UV irradiation. Theenlargement of OH groups in the 3600–3100 cm�1 range isan indication for the presence of photodegradative reac-tions that occurred in the VER–ALS network sample. Thecurves in Fig. 5 show modification of the OH groups fromVER etalon and VER–ALS semi-IPN samples.

A fast increase of the OH groups in the first 80–100 h ofirradiation can be noted, which then slows down and tendsto a plateau level. Also, the presence of ALS in the VERnetwork structure diminishes the oxidative degradation ofthe vinyl ester matrix to a certain extent. The slowing downof the VER degradation is more advanced for the sampleswith a high concentration of ALS.

The photo-oxidation of the VER–ALS semi-IPNs underUV radiation leads to unsaturated compounds withcarbonyl groups in their structures. The variation of thecarbonyl absorbances versus the irradiation time for theVER etalon sample and VER–ALS semi-IPNs (VER–ALS-2,VER–ALS-3 and VER–ALS-4) is shown in Fig. 6.

0 50 100 150 200

1,0

1,1

1,2

1,3

1,4

1,5

1,6

1,7

St/S

0

Irradiation time (h)

Fig. 5. Modification of OH groups: content on UV irradiation of: VER (-),VER–ALS-1 (C), VER–ALS-2 (:), VER–ALS-3 (;) and VER–ALS-4 (,).

100 120 140 160 180 2200 20 40 60 80 200 240

1,0

1,2

1,4

1,6

1,8

2,0

2,2S

t/S

0

Irradiation time (h)

Fig. 6. Normalized absorption of the carbonyl band versus the irradiationtime: VER (-), VER–ALS-1 (B), VER–ALS-2 (:), VER–ALS-3 (D) and VER–ALS-4 (;).

L. Rosu et al. / Polymer Testing 28 (2009) 296–300300

The change of the carbonyl absorption signals wasevaluated as a measure of the photodegradation rate. Theratio of the peak area of the studied samples at differentexposure times (S) against the surface of the peak area ofthe non-irradiated sample (S0) was calculated. Fig. 6reflects the increase of the carbonyl absorbance, especiallyin the first 70 h of UV exposure time. This behaviour can bedue to the presence of some photosensitive groups withperoxide structures in the sample mass. These structurescan result by photochemical scission of the hydroperoxides[21–23]. From this, a slight effect of photostabilization ofthe VER matrix due to ALS compound can be noted.

4. Conclusions

Some semi-IPNs based on VER and ALS were obtainedby cross-linking of VER initiated by UV radiation. Thesynthesized VER–ALS semi-IPNs were analyzed using FT-IR,DSC and OM techniques.

As a result of the cross-linking reaction, the vibrationswhich are characteristic of the double bands disappeared.The presence of a single Tg value for the VER–ALS semi-IPNsis evidence of good compatibility of the components withinthe studied networks. The presence of some functionalgroups, both in VER and in ALS, leads to possible inter-molecular hydrogen bond interactions between the polarstructures of ALS (OH, carbonyl groups) and secondaryhydroxyls from VER.

The UV radiation (l > 300 nm) led to important chem-ical modifications of the VER–ALS semi-IPNs structure.Under UV radiation, the synthesized networks undergoa photo-oxidative degradation, characterized by the pres-ence of unsaturated compounds with both hydroxyl andcarbonyl groups in their structures. A slight effect of

photostabilization of VER matrix due to the presence of ALSwas noted.

References

[1] J.R. Brown, Z. Mathys, Reinforcement and matrix effects on thecombustion properties of glass reinforced polymer composites,Composites Part A 28 (1997) 675.

[2] A. Pirvu, A.D.J. Gardner, R. Lopez-Amido, Carbon fiber–vinyl estercomposite reinforcement of wood using the VARTM/SCRIMP fabri-cation process, Composites Part A 5 (2004) 1257.

[3] A.L. Nazareth da Silva, S.C.S. Teixeira, A.C.C. Widal, F.M.B. Coutinho,Mechanical properties of polymer composites based on commercialepoxy vinyl ester resin and glass fiber, Polym. Test. 20 (2001) 895.

[4] W.M. Karnbhari, Y.W. Chin, D. Hunston, B. Benmokrane, T.Juska, R. Morgan, J.J. Lesko, U. Sorathia, D. Reynand, Durabilitygap analysis for fiber-reinforced composites in civil infrastruc-ture, J. Compos. Constr. 7 (2003) 238.

[5] H.Y. Yeh, S.C. Yang, Building of a composite transmission tower,J. Reinf. Plast. Compos. 16 (1997) 414.

[6] B. Sun, B. Gu, X. Ding, Compressive behavior of 3-D angle-interlockwoven fabric composites at various strain rates, Polym. Test. 24(2005) 447.

[7] D. Rosu, L. Rosu, C.N. Cascaval, Effect of ultraviolet radiation on vinylester network based on bisphenol A, J. Photochem. Photobiol. A:Chem. 194 (2008) 275.

[8] J.S. Wang, R.S. Manley, D. Feldman, Synthetic-polymer lignincopolymers and blends, Prog. Polym. Sci. 17 (1992) 611.

[9] H.-S. Yang, M.P. Wolcott, H.-S. Kim, S. Kim, H.-J. Kim, Properties oflignocellulosic material filled polypropylene bio-composites madewith different manufacturing processes, Polym. Test. 25 (2006) 668.

[10] D. Klempner, L.H. Sperling, L.A. Utracki, Interpenetrating PolymerNetworks, American Chemical Society, Washington, DC, 1994.

[11] C.N. Cascaval, D. Rosu, L. Rosu, C. Ciobanu, Thermal degradation ofsemi-interpenetrating polymer networks based on polyurethaneand epoxy maleate of bisphenol A, Polym. Test. 22 (2003) 45.

[12] O.L.S. Alsina, L.H. de Carvalho, F.G. Ramos Filho, J.R.M. d’Almeida,Thermal properties of hybrid lignocellulosic fabric-reinforced poly-ester matrix composites, Polym. Test. 24 (2005) 81.

[13] B. Kosikova, V. Demianova, M. Kacurakova, Sulfur-free lignins ascomposites of polypropylene films, J. Appl. Polym. Sci. 47 (1993)1065.

[14] B. Puciariello, V. Villani, C. Bonini, M. D’Auria, T. Vetere, Physicalproperties of straw lignin-based polymer blends, Polymer 45 (2004)4159.

[15] N.S. Allen, Polymer photochemistry, in: Iain Dunkin (Ed.), Photo-chemistry, vol. 35, 2005, p. 206 (Chapter 7).

[16] L. Rosu, C.N. Cascaval, D. Rosu, Curing of vinyl ester resins. Rheo-logical behaviour, J. Opt. Adv. Mater. 8 (2006) 690.

[17] L.H. Sperling, Interpenetrating Polymer Networks and RelatedMaterials, American Chemical Society, New York, 1981.

[18] O. Oblabisi, L.M. Robenson, M.T. Shaw, Polymer–Polymer Miscibility,Academic Press, New York, 1979.

[19] A.W. Signor, M.R. Van Landingham, J.W. Chin, Effects of ultravioletradiation exposure on vinyl ester resins: characterization of chem-ical, physical and mechanical damage, Polym. Degrad. Stab. 79(2003) 359.

[20] D. Rosu, C. Ciobanu, C.N. Cascaval, Polyhydroxyacrylate–poly-urethane semi-interpenetrating polymer networks, Eur. Polym. J. 37(2001) 587.

[21] D. Allard, M. Fontanille, R.E. Prud’homme, Synthesis and prop-erties of miscible epoxy/acrylic interpenetrating polymernetworks, J. Polym. Sci. Polym. Chem. Ed. 22 (1984) 3827.

[22] J. Luki, B. Ranby, Photo-oxidation of polystyrenedpart 2: formationof carbonyl groups in photo-oxidised polystyrene, Polym. Degrad.Stab. 1 (1979) 165.

[23] J. Lucki, J.F. Rabek, B. Ranby, C. Ekstr}om, Photolysis of polyesterspoly(propylene-1,2-maleate) and poly(propylene-1,2-ortho-phtha-late), Eur. Polym. J. 170 (1981) 919.