Preview only show first 10 pages with watermark. For full document please download

Study Of The Silane Modification Of Magnesium Hydroxide And Their Effects On The Flame Retardant And Tensile Properties Of High Density Polyethylene Nanocomposites

Study of the Silane Modification of Magnesium Hydroxide and Their Effects on the Flame Retardant and Tensile Properties of High Density Polyethylene Nanocomposites




  Study of the Silane Modification of MagnesiumHydroxide and Their Effects on the Flame Retardantand Tensile Properties of High Density PolyethyleneNanocomposites E.N. Cabrera-   Alvarez, L.F. Ramos-deValle, S. S  anchez-Valdes, A. Candia-Garc  ıa, F. Soriano-Corral,E. Ram  ırez-Vargas, M.C. Ibarra-Alonso, P. Pati ~ no-Soto Centro de Investigaci   on en Qu   ımica Aplicada (CIQA), Polymer Processing Dept., Boulevard Enrique Reyna,No. 140, Saltillo, Coahuila 25294, Mexico The modification of magnesium hydroxide (MH) withtriethoxy vinyl silane (TVS) was carried out via threedifferent methods and the results are discussed withrespect to their effect on the flame retardant and thetensile properties of high density polyethylene (HDPE)nanocomposites. Via a xylene suspension of MH andTVS, via a water suspension of MH and TVS, and viapowder mixing of MH and TVS. It was found that in allthree cases, Si A O A Mg bonds formed on the MH par-ticles surface. Also, this silane modification induced acertain level of particle agglomeration, but withoutmodifying their morphology or particle size. It wasalso found that the flame retardant properties of theHDPE nanocomposites did not improve because ofthe use of silane modified MH. During the flameretardant tests, all nanocomposites passed the UL-94-HB, but it was observed that the flame permanencetime was longer when using modified MH. The tensileproperties were negatively affected by the addition ofunmodified MH; the nanocomposites became hardand brittle, with reduced flexibility. This negative effectwas diminished when using silane modified MH.POLYM. COMPOS., 00:000–000, 2013.  V C  2013 Society of Plastics Engineers INTRODUCTION Polyolefins, such as polyethylene (PE) are used inmany different applications, due to their good all-aroundproperties. Polyolefins, however, have a serious limitationwhen flame resistance is of importance in a given appli-cation. To offset this limitation, flame retardants areadded into the PE composition. Some efficient halogen-free flame retardants are the hydrated metallic hydroxides,such as magnesium and aluminum hydroxides [1–4].These metallic hydroxides reduce the flammability of polyolefin compositions, though high hydroxide contenthas to be used in order to attain acceptable flame retard-ancy. This high filler content has a negative effect onseveral mechanical properties, turning the PE into a brit-tle and fragile composition with very poor tensileproperties.This negative effect is mainly due to the lack of inter-actions and/or compatibility between the polymer matrixand the filler particles. One way to promote these inter-actions and/or compatibility is through the addition of acoupling agent, such as a silane or a titanate [5–7]. In thisrespect, the organic silanes of the R A Si A  (OR) 3  typehave shown to be very effective for the surface modifica-tion of the magnesium and aluminum hydroxides, gener-ating  A Si A O A  metal bonds, which force the inter-actionsbetween the polymer matrix and the metallic hydroxideparticles [8]. However, care must be taken in order tofind the adequate reaction conditions, when adding a cou-pling agent, in order to promote the formation of thestrong covalent  A Si A O A  Metal bonds.The reaction between the R A Si A  (OR) 3  silane withmagnesium hydroxide (MH) is depicted in Fig. 1. Theethoxide groups from the silane react with the MH, pro-ducing ethanol via a condensation reaction, plusMg A O A S A  bonds. The R groups in the silane are theones responsible for the inter-action and/or compatibilitytowards the polymer matrix [9].Several methods have been used for the surface modi-fication of metallic hydroxides, one of which is due toChen et al. [10, 11] who carried out the MH modificationwith silanes and titanates, in a high speed powder mixer.They state that the coupling agent is adsorbed on thehydroxide surface, though they do not show any evidenceof Si A O A Si bonds. They do show, however, that themechanical properties of polypropylene composites with Correspondence to : L.F. Ramos-deValle; e-mail: [email protected] or S. Sanchez-Valdes; e-mail: [email protected] grant sponsor: Conacyt; contract grant number: CB-104865;contract grant sponsor: PROMEP.DOI 10.1002/pc.22753Published online in Wiley Online Library ( V C 2013 Society of Plastics Engineers POLYMER COMPOSITES—2013  Article first published online: 31 OCT 2013, DOI: 10.1002/pc.22753  modified MH were superior when compared to thosewhen using the unmodified hydroxide.Recently Wan [12] described a patent about a particlesurface modification method using ultrasound, which per-mits the modification in a short period of time. Zhang[13] reported the surface modification of magnesiumhydroxide using ultrasound. Kong et al. [14] carried outthe surface modification of MH with 5 wt% of a silane,in xylene, at 140  C for 6 h, using dibutyl tin dilaurate ascatalyst. Thereafter, they found that LLDPE compositeswith 50 wt% of modified MH attained a V-0 classifica-tion in the UL-94 flame retardancy standard. But theyalso found, however, that when using less than 5 wt% of silane during the surface modification reaction, resulted inLLDPE composites that attained a V-2 classification.While studying EVA with nano-sized MH composites,Jiao et al. [15] found that fatty acid modified MH nano-composites produced an increase in mechanical propertiesas compared to unmodified MH composites. Bahattabet al. [16] found that crosslinked LDPE/EVA compositeswith surface modified MH produced superior tensile prop-erties. Recently, it was reported [17] that the silane modi-fied MH nanoparticles markedly improved the mechanicalproperties of epoxi resin, as compared to the same epoxiresin with an unmodified MH.In general, surface modified metallic hydroxides haveshown to produce good mechanical properties such as, ten-sile strength and elongation, as well as impact resistance.From the above, it can be concluded that the studieson the MH silane modification methods have providedample information on the matter, but most of these havenot been related to the effect on the mechanical proper-ties and flame resistance of the resulting polymer composites.Therefore, the purpose of this work is to study theeffect of different MH surface modification methods onthe mechanical properties and flame resistance of HDPE/ MH nanocomposites. EXPERIMENTAL  Materials Used  Materials used were: magnesium hydroxide (MH),with nominal particle size of 15 nm, from Nanostructuredand Amorphous Materials Inc, USA; Triethoxy Vinyl Sil-ane (TVS), dibutyl tin dilaurate and boric acid, fromAldrich, USA; and the Xylene solvent from CTR Scien-tific, Mexico. The high density polyethylene used wasfrom Dow Chemical, USA, with MFI of 0.7 g/10 min,density of 0.954 g cm 2 3 , and fusion temperature ( T  m ) of 127  C. Surface Modification The MH surface modification was carried out follow-ing three different methods. Method 1.  Xylene and MH were added into a 1-Lflask in order to have 500 cm 3 of a 10 wt% suspensionof MH in xylene. The system was kept under agitationwhile the temperature was increased up to 130  C. Atthis point, 5 wt% with respect to the MH, of a 2:1TVS:Boric Acid mix, plus 0.5 mol% with respect to theMH, of dibutyl tin dilaurate catalyst [14], were added.The reaction conditions (130  C, agitation, and reflux)were maintained for 4 h, after which the system wascooled down and the reaction stopped. The product(modified MH) was purified either by filtration andwashed with ethanol MHX F  or by evaporation andwashed with ethanol MHX E . Method 2.  Distilled water and MH were added into a1-L flask in order to have 500 cm 3 of a 6 wt% suspensionof MH in water. The system was kept under agitationwhile the temperature was increased up to 90  C. Sonica-tion (19.65 MHz and 300 W) was applied to the suspen-sion for 45 min to activate the magnesium hydroxide OHgroups. At this point, 5 wt% with respect to the MH, of a2:1 TVS:Boric Acid mix was added. The reaction condi-tions (90  C and agitation) were maintained for 1 h, after which the system was cooled down and the reactionstopped. The product (modified MH) was purified either by filtration and washed with ethanol MHW F  or by evap-oration and washed with ethanol MHW E . Finally, follow-ing the same procedure, but at room temperature andwithout sonication, the ingredients (water, MH, TVS, andboric acid) were mixed for 20 min, after which, the sys-tem was kept at 80  C for 24 h in order to dry the productMHW D . Method 3.  In this case, 500 g of MH plus 5 wt% withrespect to the MH, of a 2:1 TVS:Boric Acid mix wereadded into a high speed powder mixer. The mixing wasstarted and kept at 1700 rpms, at 30  C, for 30 min. Theproduct (modified MH) was designated as MHM SB .Another mix, was prepared in the same manner but add-ing only 3.33 wt% of TVS, without boric acid. This prod-uct was designated as MHM S .All modified MH samples were powdered and driedunder vacuum, at 60  C, for 12 h. Unmodified magnesiumhydroxide was used as reference and is designated as MH.Diagram 1 summarizes the different modificationprocedures. FIG. 1. Typical reaction of triethoxy-R-silane with magnesiumhydroxide. 2 POLYMER COMPOSITES—2013 DOI 10.1002/pc   Preparation of HDPE/MH Nanocomposites Each nanocomposite simple was prepared as follows:HDPE was mixed with unmodified and modified MH in aBrabender internal mixer, using roller type rotors, at190  C and 60 rpm, for 10 min. In all cases the unmodi-fied or modified MH concentration was 55 wt%. Theresulting nanocomposites were grounded and then com-pression molded to obtain 150  3  150  3  3 mm 3 laminatesfrom which test specimens were taken.Tensile and flame retardant properties of theobtained nanocomposites were evaluated as describedbelow.  Fourier Transform Infra-Red Spectroscopy (IR) Spectroscopic analyses were carried out in a NexusFTIR spectrometer between 400 and 4,000 cm 2 1 , using32 scans and 4 cm 2 1 resolution. Thermogravimetric Analysis (TGA) Thermogravimetric analyses were performed in a TAInstruments TGAQ500 between 30 and 650  C, using aheating rate of 10  C min 2 1 , under a nitrogen flow of 50ml min 2 1 . Scanning Electron Microscopy (SEM and STEM) For the analysis of MH particles, a 10 wt% suspensionof MH in acetone was prepared and subjected to sonica-tion for 1 h. Small samples of this suspension wereplaced on copper grids for STEM analysis. SEM Analysisof HDPE/MH nanocomposites was carried out on the Au-Pl coated fractured surface of small pieces from the previ-ously obtained laminates. All samples were fractured atliquid nitrogen temperature.All samples were additionally examined through anEDAX attachment for determining the silicon content.  Flammability Tests UL 94 Flammability tests were performed according to UL-94,in both, vertical (94V) and horizontal (94HB) arrange-ments, on 125  3  13  3  3 mm 3 specimens cut fromcompression molded laminates, which were previouslyconditioned for 48 h at 23  C and 50% RH. Tensile Properties ASTM D638 Tensile tests were carried out according to ASTM D638 in an Instron 4301 tensile testing machine at a FIG. 2. IR spectra of MH; TVS; a physical mix of MH and TVS; andthe method 1 modified MHX E  and MHX F . DOI 10.1002/pc POLYMER COMPOSITES—2013 3 sample  deformation speed of 8 mm min 2 1 , using type V speci-mens of 63.5 3 13 3  3 mm 3 . RESULTS AND DISCUSSION Surface Modification of Magnesium Hydroxide (MH) Figure 2 shows the IR spectra of MH and TVS, aswell as those of the method 1 modified MH. The charac-teristic peaks of MH, at 1,421 and 1,487 cm 2 1 , due to theMg A O bonds and at 3,443 and 3,693 cm 2 1 , due to theO A H bonds of the MH, are clearly observed [18, 19].Figure 2 also shows the characteristic peaks of TVS, at1079 and 1104 cm 2 1 , due to the Si A O bonds and at2,890–3,000 cm 2 1 due to the C A H of the ethoxy groups,plus a slight signal at 1,598 cm 2 1 , which corresponds tothe C @ C of the vinyl groups.A 95/5 MH/TVS mix, used as reference, was preparedand grinded in a mortar and designated as MH-TVS MIX.Figure 2 shows its spectrogram, where the characteristicpeaks of the Mg A O groups in MH, at 1,421 and 1,487cm 2 1 and of Si A O groups in TVS, at 1,079 and 1,104cm 2 1 , are clearly observed.Following method 1, when the xylene eliminationwas by means of evaporation (MHX E ), two new signalsappeared around 1,044–1,126 cm 2 1 which correspond tothe siloxane groups (Si A O A Si) [19] formed on the MHparticles surface due to the modification reactions.Nonetheless, these peak could also be due to the silanegroups of the unreacted TVS, since there are still visiblepeaks at 2,855–2,958 cm 2 1 , and at 3,022–3,060 cm 2 1 ,which correspond to the aliphatic CH 3 A CH 2 A O A groups and the vinyl CH @ CH groups of the TVS,respectively. This implies that the condensation reactionwas not “complete,” since the unreacted TVS is noteliminated by the evaporation process, and therefore, theTVS aliphatic and vinyl group signals still appear on thespectrogram. When purification is carried out by filtra-tion, the unreacted TVS is eliminated along with thesolvent. In addition, there is a signal at 1,263–1,283cm 2 1 due to the O A H groups from the boric acid andfrom the MH.When the xylene elimination was by means of filtra-tion (MHX F ), again, there appears the peak at 1,044cm 2 1 , due to the siloxane groups (Si A O A Si) [19] formedon the MH particles surface, however, the peaks at2,900–3,000 cm 2 1 , due to the aliphatic CH3 A CH2 A O A groups of the TVS are not observed anymore.From the above, it can be concluded that the TVS isclearly reacting with the MH, generating Si A O A Si andSi A O A Mg groups.Figure 3 shows the IR spectra of the method 2 modi-fied MH. Also, when the water elimination was by meansof evaporation (MHW E ), a new signal appears around1,033–1,066 cm 2 1 which corresponds to the Si A O A Sigroups formed on the MH particles surface due to themodification reactions [19]. Nonetheless, these peak couldalso be due to the silane groups of the unreacted TVS,since there are still visible peaks at 2,855–2,958 cm 2 1 ,and at 3,022–3,060 cm 2 1 , which correspond to the ali-phatic CH3 A CH2 A O A  groups and the vinyl CH @ CHgroups of the TVS, respectively.This means that the MH modification was incompleteand there is still TVS without reacting, since it remains inthe composition when the purification is carried out byevaporation. Also, there appears a signal at 1,263–1,283cm 2 1 due to the O A H bond in MH and in boric acid.When the water is eliminated by filtration (MHW F ),again, there appears the peak at 1,040 cm 2 1 , due to thesiloxane groups (Si A O A Si) [19] formed on the MH par-ticles surface, however, the peaks at 2,855–3,060 cm 2 1 ,due to the aliphatic CH3 A CH2 A O A  groups of the TVSare not observed anymore, that is, the unreacted TVS iseliminated along with the solvent.On the other hand, when the water is eliminated bydrying at 80  C for 24 h, there also appears a peak at1,040 cm 2 1 due to the Si A O A Si and at 1,263–1,283cm 2 1 due to the boric acid. The peaks corresponding tothe aliphatic CH3 A CH2 A O A  groups and the vinylCH @ CH groups of the TVS are no longer there, since theunreacted TVS was eliminated by the drying process.From the above, it can also be concluded that the TVS isclearly reacting with the MH, generating Si A O A Si andSi A O A Mg groups.Figure 4 shows the IR spectra of the method 3 modi-fied MH. Either including (MHM SB ) or not (MHM S ) theboric acid, in both cases the spectrograms show the char-acteristic peaks of MH, at 1,421 and 1,487 cm 2 1 , due tothe Mg A O bonds and at 3,443 and 3,693 cm 2 1 , due tothe O A H bonds of the MH [13, 18, 19].In addition, the signal of TVS, at 1,045 cm 2 1 , due tothe Si A O bonds and at 2,900–3,000 cm 2 1 due to theC A H of the ethoxy groups. When boric acid was FIG. 3. IR spectra of MH; and the method 2 modified MHW E ,MHW F , and MHW D . 4 POLYMER COMPOSITES—2013 DOI 10.1002/pc  included (MHM SB ), a signal at 1,268 cm 2 1 , due to theacid O A H groups was observed.Deconvolution of the MHM SB  compound IR spectrumbetween 2,000 and 1,500 cm 2 1 , in Fig. 4A, permits amuch clearer view of the vinyl group signal at 1,627cm 2 1 , which was overlapped with the MH signal. Thiscorroborates the presence of the vinyl group in the modi-fied MH.It is worth mentioning that the compounds designatedas MHM SB , as well as those designated as MHX F ,MHW F  y MHW D , they all present a signal at 1,033– 1,066 cm 2 1 , which corresponds to the Si A O A Si groups,independently of the modification method used. Also, itcan be observed that none of these nanoparticles presentthe TVS aliphatic group signals, which was, partially con-sumed through the modification reaction or partially fil-tered in the purification step.To confirm the condensation reaction between the MHand the TVS taking place, a model compound was pre-pared following method 1, with purification by filtration,but using 40 wt% TVS with respect to the MH(MHX F 40). This, along with MH and TVS were analyzedvia TGA, and the results are presented in Fig. 5.Taking as reference point the one at 15 wt% loss, itcan be observed that TVS reached the 15 wt% loss at56  C and the whole 100 wt% loss at 99  C (TVS is a veryvolatile liquid with flash point of 34  C). The unmodifiedMH on the other hand, reached the 15 wt% loss at367  C. At 400  C, the MH reached about 30 wt% loss, bydecomposing into H 2 O and MgO [11].The modified MH, on the other hand, presents a 15wt% loss at a much higher temperature, namely 411  C.This is assumed to be due to the formation of the stronger Si A O A Mg bonds. If there were no reaction between TVSand MH, there would be two weight loss steps, one corre-sponding to TVS and another corresponding to MH. Thatis, the behavior of MHX F 40 is not that of a mix of twodifferent substances, but that of a new single substance,generated by the reaction between TVS and MH. Finally,it can be observed that at 450  C, the combustion of MHproduces   70 wt% of an inorganic residue consisting of MgO, whereas the combustion of MHX F 40 produces   75wt% of an inorganic residue consisting of MgO and SiO 2 .The effect of the modification method upon the modi-fied MH particle size and morphology was assessedthrough STEM, and Fig. 6 shows the results. Figure 6Ashows that the unmodified MH particles have a pseudo-spheric morphology with a particle size around of 144nm. When modified according to either method 1 (Fig.6B), 2 (Fig. 6C), or 3 (Fig. 6D), it can be observed thatin general, the pseudospheric morphology and the particlesize are not altered, but also, it is observed that the par-ticles tend to agglomerate, independently of the method FIG. 4. (A) IR spectra of MH; and the method 3 modified MHM S  and MHM SB ; and (B) Deconvoluted IRspectra of MHM SB  between 2,000 and 1,500, clearly showing the peak at 1,627 cm 2 1 .FIG. 5. Thermogravimetric analysis of MH, TVS, and MHX F  40%. DOI 10.1002/pc POLYMER COMPOSITES—2013 5