Synthesis and characterization of ozonized polyethylene.
J.J. Robin (1)(*)
N. Torres (2)
J. Casteil (1)
Polyolefins (LDPE and HDPE) can be ozonized with ozone/air gas flow to create peroxide and hydroperoxide groups on polymers. These reactive groups can be used to initiate radical polymerization of vinylic monomers and produce graft copolymers. In this study, the optimization of the experimental conditions and the explanation of the phenomena occurring during the ozonization of polyolefins in a fluidized bed have been studied. The most reproducible methods for placing peroxide and hydroperoxide groups onto polyolefins are the iodometric method and indirect titration by thiol. This work shows that it is necessary to control the temperature of the fluidized bed to avoid the acceleration of the reaction and the degradation of the polymers. Then, we studied the effect of different parameters such as the crystallinity and the granulometry of polymers, the time of ozonization and the effect of the load of the reactor on the quality of ozonization. Infrared spectroscopy and steric exclusion chromatography were used t o characterize the ozonized polymers. We show that double bonds of PE, particularly vinyl groups, are very reactive to ozone and that ozonized HDPE leads essentially to the formation of hydroperoxide groups, whereas ozonized LDPE leads to the formation of peroxide groups.
During recent years, the chemical modification of polymers has undergone strong development, with a view to conferring new properties upon these materials. For example, several studies have concerned the improvement of characteristics such as barrier properties, the wetting or the resistivity of macromolecules. Among the polymers tested, polyethylene (PE), polypropylene (PP) and their copolymers such as ethylene-propylene-diene monomer (EPDM), ethylene-propylene rubber (EPR) and ethylene-vinyl acetate copolymer (EVA) have been most studied. To modify the surface of these polymers, the methods used have been oxidation with flame, corona treatment, or oxidation with concentrated mineral acids. These processes lead to a superficial oxidation of material and entail the formation of alcohol or acid groups (1). These treatments, interesting for industrial development, are insufficient to modify the materials in depth and lead to a true modification of all the macromolecules. Researchers are moving toward more effic ient solutions for treating the "heart" of materials. It has been reported that treatments by electron beams or irradiation (2-4) lead, first, to the formation of free radicals by hydrogen abstraction, and second, to the formation of peroxide and hydroperoxide groups in the presence of oxygen molecules. These methods using radiation are very efficient but also very expensive and not easy to apply. Researchers are therefore moving toward other processes, such as the activation of a polymer by ozone. Several studies have shown that polymers such as PP (5, 6), LDPE (7, 8), poly(vinylidene-fluoride) (PVDF) (9, 10), EVA (11, 12), and poly(vinylchloride) (PVC) (13) can be ozonized with air/ozone gas flow to create peroxide and hydroperoxide groups by pulling hydrogen atoms out of polymer chains (14). More recently, other works have concerned the chemical modification of polymers by radical grafting of monomers. This process consists of introducing a peroxide as an initiator in the polymer/monomer system. The most-s tudied monomers have been maleic anhydride (15-17), vinylic monomers (18), or acrylic and methacrylic monomers (19-22).
The purpose of this work was to study the last two ways described, i.e., ozonization of polymers and chemical grafting, because the formation of peroxide and hydroperoxide sites on the polymer chains leads to grafting of monomers, avoiding the use of commercial peroxides. Ozonization treatment is very interesting since it allows the modification of polymers by formation of substances such as alcohol or acid by decomposition of peroxide and hydroperoxide groups and the grafting of monomers. Most of the works have concerned the use of macroinitiators formed during the ozonization rather than the explanation of phenomena or the study of parameters that control this process. In this study, we characterized the active substances obtained during ozonization. This treatment is achieved on polymers presenting different chemical structures and using different experimental conditions.
HDPE (HD 6621, Stamylan) and LDPE (LD 1922Z, Stamylan) were supplied by DSM. The powdered polyethylenes were sifted before being activated through an ozone current. For example, an LDPE powder that slips through a sifter of 600 [micro]m but is retained by a sifter of 200 [micro]m is noted LDPE 200-600. All other chemicals were purchased from Aldrich.
Experimental Conditions of Ozonization
The reactor of ozonization or the column of fluidization is made up of two cylindrical tubes of 50-mm and 100-mm diameters. They are linked by a truncate cone 100 mm high. The distributor of the gas is composed of a thin cloth of stainless steel surmounted by a uniform layer of glass balls (the thickness of the layer is equal to 1.7 cm and the diameter of a single ball is equal to 1.5mm). At the top of the reactor, a collector collects excess ozone and oxygen. The gaseous mixture is injected at the base of reactor under the gas distributor. This mixture is heated by spiral resistance of 1500 W. The heating power and the regulation are ensured by a microprocessor. The main parameters to consider are:
* the temperature measured in the fluldized bed, [T.sub.R],
* the gaseous volumic throughput. [Q.sub.u],
* the ozone rate in the gaseous mixture at the entrance of the column, [Y.sub.03],
* the time, t,
* the pressure, p, in the distributor and the loss of total load through the distributor and the fluidized bed, [[DELTA].sub.p],
* the temperature of fluidization gas in the distributor, [T.sub.D].
The ozone generator (Trailigaz Minibloc 76 Model) releases a ozone current of 16 g/l at the most. The ozone concentration generated is adjustable as a function of the electric power supplied to the equipment.
Methods of Titration
Titration by DPPH
One tenth of a gram (0.1 g) of ozonized PE was dissolved in a solution (10 mL) of 2,2-diphenyl-1-picryl-hydrazyl (DPPH) (0.6 g/L) in redistilled o-xylene. A blank solution without polymer was made up under the same conditions. The solutions were saturated with argon for 45 minutes and the reactors were maintained in a thermostated oil bath at 110[degrees]C for 10 minutes. The reactors were quickly cooled with cold water for 5 minutes before the solutions were diluted with 90 mL of pure isopropanol. From the blank solution and by successive dilutions with o-xylene/isopropanol (10/90), we plotted a calibration curve giving the variation in the absorbance of the solution at 520 nm as a function of its concentration. The solution containing ozonized PE was filtered through a fine filter (0.2 [micro]m porosity) before the concentration in residual DPPH was determined.
Indirect Titration by Thiol
One gram (1 g) of ozonized PE and 0.1 g of [C.sub.8][F.sub.17][H.sub.4]SH were dissolved in 50 mL of o-dichlorobenzene. The solution was saturated with argon for 30 minutes and was placed in a thermostated oil bath at 130[degrees]C. The solution was quickly cooled with cold water for 5 minutes, and then 10 mL of ethanol were introduced prior to titrate residual thiol with 0.03 N iodine solution in the ethanol.
One gram (1 g) of ozonized PE was immersed in decalin (30 mL) to swell. Sodium iodide, 10 mL, in an isopropyl alcohol solution (20 g/L) and 2 mL of acetic acid were added. The solution was saturated with argon for 15 minutes and then was placed in a thermostated oil bath at 100[degrees]C for 10 minutes. With these experimental conditions, LDPE is entirely soluble, whereas HDPE is not entirely soluble. After cooling with cold water for 5 minutes, deionized water (10 mL) was added and the solution was titrated with N/100 sodium thiosulfate. The titrations were always repeated two or three times.
Infrared spectra were recorded on a Fourier Transform Spectrometer (Nicolet 510P) from films obtained under pressure (190[degrees]C, 2 min, 150 bars).
The weight and number-average masses ([M.sub.w] and [M.sub.n]) were determined by steric exclusion chromatography with columns PL GEL MIXED B. The samples were dissolved at 145[degrees]C in 1,2,4-trichlorobenzene. The standardization was performed with standards of PE. The polymolecularity index ([M.sub.w]/[M.sub.n]) gives information on molecular mass distribution.
RESULTS AND DISCUSSION
The radical mechanism describing the ozonization of polyethylene is given by Kefeli et al. (14):
RH + [O.sub.3] [right arrow] [RO.sub.2.sup.*] + [OH.sup.*]
[RO.sub.2.sup.*] + RH [right arrow] [R.sup.*] + [RO.sub.2]H
[R.sup.*] + [O.sub.2] [right arrow] [RO.sub.2.sup.*]
[RO.sub.2.sup.*] + [R.sup.*] [right arrow] [RO.sub.2]R
This mechanism is very interesting, though it does not take into account the polymer structure. Indeed, it is well known that ozone reacts strongly with the unsaturations of polyethylene to form unstable ozonides. Then, the decomposition of ozonides produces inert substances such as alcohol, ketone or acid. It is necessary to take into account the rate of branching because the ozone reacts preferentially with tertiary carbons. In this study, we quantified the behavior with ozone of two polymers obtained by different processes (Ziegler or Philips). The peroxide content is not easy to determine because there are very few functions on the polymer backbones, and they are also difficult to quantify. Moreover, the polymers studied are only slightly soluble in the classical solvents used in organic chemistry. To determine the peroxide content, it has been necessary to perfect a reliable method. Several researchers have already tackled this problem.
Colorimetric Titration by Diphenyl Picrylhydrazyl (DPPH)
This method detailed by Boutevin et al. (7) has been used to determine peroxide and hydroperoxide group content on PE (8, 13), PVDF and EVA (11, 12). This method supposes that free radicals such as DPPH (Fig. 1) react with radicals coming from the decomposition at 110[degrees]C of peroxide and hydroperoxide groups onto ozonized PE and that one DPPH mole neutralizes one mole of free radicals. After 10 minutes at 110[degrees]C, the DPPH concentration remains constant, indicating that creation of free radicals in the solution has ceased. Then, the residual DPPH concentration is determined by colorimetry at 520 nm. The absorbance varying linearly with the concentration (Lambert-Beer's low), we thus obtain the concentration in residual DPPH. In practice, for every titration, a blank determination is done in the same conditions without ozonized PE and from different dilutions to plot a calibration curve. This curve gives the absorbance at 520 nm as a function of the concentration in DPPH solution ([Ab.sub.o] = f(c)). Before achieving the titration by colorimetry, the solution containing ozonized PE is precipitated into pure iso-propanol and filtered. The peroxide and hydroperoxide content, [T.sub.oo] (mol of eq/g of polymer), is equal to half of the concentration in DPPH radicals (also called active oxygen rate [T.sub.o]) that have reacted, since each peroxide or hydroperoxide group produces two free radicals upon decomposition. The peroxide and hydroperoxide rate is given by Eq 1:
[T.sub.oo] = [T.sub.ROOH] + [T.sub.ROOR] = [T.sub.o]/2 = ([C.sub.o] - [C.sub.PE])*V/2*[m.sub.PE].100 (1)
where [T.sub.o] = active oxygen rate, [C.sub.o] = concentration in DPPH of standard solution (mol/L), [C.sub.PE] = concentration in DPPH of solution containing ozonized PE (mol/L), V = total volume of the solution (mL), [m.sub.PE] = mass of PE (g).
The determination of peroxide and hydroperoxide groups on ozonized PE by DPPH titration gives non-reproducible results ([+ or -]50%). More recently, Brondino (23) also obtained nonreproducible results on ozonized PP and ozonized PVDF in mass. The nonreproducibility of this method was not pointed out by Robin (13), Sarraf (8) or Serdani (10). Fargere (11, 12) showed that the DPPH method can be used to determine the amount of peroxide and hydroperoxide groups on ozonized EVA. Fargere considered that the majority (>95%) of hydroperoxide groups react quickly, whereas peroxide groups react slowly and moreover are responsible for the rupture of chains. He noticed that ozonized EVA can be considered a macroinitiator since it contains essentially hydroperoxide groups, which can initiate radical polymerization. To explain the nonreproducibility of titration by DPPH of ozonized polymers in mass, a possible hypothesis is that the amount of sample tested (0.1 g) is too small since the ozonization in mass is less homogene ous than the ozonization in solution. To verify this hypothesis, we could achieve titration with 1 g of sample. But, that would lead to very important volumes of reagent and modify the kinetics of the reaction because the increase in temperature will be slower. The DPPH method cannot be used to determine hydroperoxide and peroxide groups on ozonized polymers because the results obtained are not reproducible.
Iodometric titration is a general method to determine peroxide and hydroperoxide groups on polymers. This method, described in the literature by Wagner et al. (24), is based on the oxidation of iodine ions by peroxide species in an acid medium. [I.sub.2] is titrated by a solution of sodium thiosulfate (2 [Na.sup.+] + [S.sub.2][O.sub.3.sup.2-]) as follows:
ROOR' + 2[I.sup.-] + 2[H.sup.+][[right arrow].sup.[DELTA]] [I.sub.2] + ROH + R'OH
[I.sub.2] + 2 [S.sub.2][O.sub.3.sup.2-] [right arrow] 2[I.sup.-] + [S.sub.4][O.sub.6.sup.2-]
Peroxide rate is given by Eq 2.
[T.sub.oo]= [N.sub.[S.sub.2][O.sup.2-.sub.3]] * [V.sub[S.sub.2][O.sup.2-.sub.3]]/2.[m.sub.PE] (mol of eq/g of polymer) (2)
To determine the peroxide groups on polymers of low solubility, an iodine ion is generally added in the reactor in the form of an isopropanol solution saturated in KI ([C.sub.KI] [less than or equal to] 4 g/1 at 20[degrees]C). However, to titrate all peroxide groups onto ozonized PE, an excess of iodine ions is necessary. So, we used an isopropanol solution with NaI at 20 g/1 (saturation: 180 g/1 at 20[degrees]C). To achieve the titration and determine the time of reaction leading to the total decomposition of peroxide groups onto ozonized PE, we used dibenzoyl peroxide since it possesses a similar half lifetime ([t.sub.1/2] = 28 min at 100[degrees]C) in comparison with those of peroxide species coming from ozonized PE (8) ([t.sub.1/2] = 5 mm at 100[degrees]C). We noted that dibenzoyl peroxide is entirely titrated after 10 minutes, giving a half lifetime in the presence of [I.sup.-] at 100[degrees]C of about one minute. In practice, a standard titration is achieved with non-ozonized PE because It is known that Iodi ne can add up onto the unsaturations of polyolefins (25). The amount of peroxide species is given by the difference between the values obtained when titrating ozonized PE and those obtained when titrating standard solution. For six titrations achieved on ozonized HDPE or LDPE, we found reproducible results with a deviation of 0.1 X [10.sup.-5] eq/g.
Indirect Titration by Thiol
At 130[degrees]C, radicals coming from the decomposition of peroxide and hydroperoxide species are created. The addition of a very efficient transfer agent ([C.sub.8][F.sub.17][C.sub.2][H.sub.4]SH named [R.sub.F]SH) leads to the stabilization of peroxide radicals. The reactional mechanism is described as follows:
ROOR [[right arrow].sup.[DELTA]] 2[RO.sup.*]
[RO.sup.*] + [R.sub.F]SH [right arrow] ROH + [R.sub.F][S.sup.*]
[R.sub.F][S.sup.*] + [R.sub.F][S.sup.*] [right arrow] [R.sub.F][SSR.sub.F]
Residual thiol is then titrated with a Iodine solution:
2[R.sub.F]SH + [I.sub.2] [right arrow] [R.sub.F][SSR.sub.F] + 2HI
[T.sub.oo] = 1/2.[m.sub.PE] * [[m.sub.thiol]/[M.sub.thiol] - [V.sub.[I.sub.2]]*[N.sub.[I.sub.2]]] (mol of eq/g of polymer) (3)
where [m.sub.PE] = mass of PE, [m.sub.thiol] = mass of thiol, [M.sub.thiol] = molar mass of thiol, [V.sub.[I.sub.2]] = volume of iodine solution, [N.sub.[I.sub.2]] = normality of iodine solution ([N.sub.[I.sub.2]] = 2.[M.sub.[I.sub.2]] with [M.sub.[I.sub.2]] = molarity).
Thiols can react on the double and banging bonds of a polymer since it has been shown by Boutevin et al. (26) that 1-2 and 1-4 double bonds of polybutadiene can be attacked by thiols.
[FORMULA NOT REPRODUCIBLE IN ASCII]
Then, we verified the stability of thiol, plotting the evolution of the amount of thiol present in the medium as a function of time (Fig. 2). We noted that [R.sub.F]SH is stable during 60 minutes at 100[degrees]C. In the presence of ozonized or non-ozonized PE, we observed after 5 minutes that the ratio [[[R.sub.F]SH]/[R.sub.F]SH].sub.0] is equal to 110%. To reach the colorimetric turning, more than 10% of iodine must be added. That could be because:
* iodine reacts faster than thiol onto PE unsaturations. Indeed, iodine is often used to titrate PE unsaturations (26, 27),
* the determination of equivalent point by colorimetric method with precipitated PE is perhaps executed with a delay compared to that without PE.
The graphic curve representing the rate of residual thiol titrated with iodine for virgin HDPE shows a consumption of thiol coming from double bonds of HDPE. This polymer obtained by the Philips process possesses more unsaturations than a polymer obtained from the Ziegler process (Table 1) (28). An iodometric titration achieved on the double bonds of HDPE Philips leads to a rate of double bonds equal to 8.5 X [10.sup.-5] (C = C)/g (Table 2). This rate represents approximately the thiol consumption of virgin HDPE. Figure 2 shows that after 30 minutes, the consumption of thiol is more important for virgin HDPE. On the other hand, after 50 minutes, this consumption is the same. This shows that the peroxide species have totally reacted with thiol after 30 minutes, and that they react faster with thiol than with double bonds. Moreover, ozone has attacked a large part of double bonds of PE to form a different type of species. To determine the peroxide species on ozonized HDPE by this method, one must take into acc ount the difference between the rates of residual thiol obtained with ozonized HDPE and those obtained with virgin HDPE. The difference of reactivity that exists between the peroxides and the double bonds towards thiol between 20 and 30 minutes gives an idea of the peroxide rate. In this case, [T.sub.[infinity]] is equal to 5 X [10.sup.-5] eq/g. To compare the results obtained, we used the iodometric method to titrate this sample. We found that [T.sub.[infinity]] is equal to 5 X [10.sup.-15] eq/g. This result is in agreement with those obtained by indirect titration using thiol.
Experimental Conditions of the Ozonization Reaction
The methods of Robin (9, 13), Sarraf (7, 8), Serdani (10) and Brondino (23) were carried out in a reactor equipped with a tubular fluidized bed. The working temperature was obtained with thermostated oil flow in a double envelope. With this device, the temperature inside the powder was not homogeneous. On the other hand, in this study the device has been improved since the air/ozone gas flow is controlled in temperature.
However, the ozonization of polymers being an exothermic reaction, we underlined the differences of behavior between HDPE and LDPE treated under the same experimental conditions but without thermal regulation. We observed that the exothermicity of the reaction is distributed in the fluidized bed as long as the polymer did not reach a critical temperature. When this temperature is reached, there is acceleration of reaction, melting and coalescence of grains in the fluidized bed.
Figure 3 gives the temperature measured inside HDPE and LDPE samples as a function of reaction time. The Ozonization of polymers was carried out without thermal control, and with heating power of 40 W and a starting temperature of ozone production equal to 50[degrees]C. When the ozone production starts, the reactor temperature increases suddenly by 2[degrees]C before stabilization. After a few minutes, the ozonization reaction runs away, leading to a strong and regular increase in temperature ([approximately equal to] 2[degrees]C/min). The fluidization of polymer becomes more and more difficult because there is coalescence of particles. Then, the temperature increases by 10[degrees]C/min, leading to the melting of polymer. At 100[degrees]C, the signal of the [T.sub.R] measurement is saturated. The ozonization of HDPE starts at 48[degrees]C, which leads to an increase in the temperature of 2[degrees]C. After a few minutes, the HDPE temperature rises by 0.75[degrees]C/min; then the exothermicity of reaction becomes faster and the te mperature increases by 10[degrees]C/min, leading to the melting of polymer and the coalescence of particles. The powders of HDPE and LDPE react differently to the ozone. LDPE being more sensitive to ozone than HDPE, there is a more rapid acceleration of the reaction. However, we have observed that HDPE reacts very violently at 64[degrees]C, whereas the acceleration of the LDPE reaction occurs at about 80[degrees]C.
To control the acceleration of ozonization reaction, we installed a thermal regulation of the temperature of the fluidized bed. Figure 4 shows the results obtained in presence or in absence of thermal control. We note that there is stabilization and control of the temperature around the order temperature when HDPE ozonization occurs with thermal control.
Then, we studied the optimal ozonization conditions of HDPE and LDPE powders of the same granulometry (Figs. 5 and 6). These Figures show that the volumic throughput and so the superficial speed of fluidizer gas (reactive mixture air/ozone) are very important parameters that have an effect on the quality of the fluidization and on the agglomeration of particles. We must take into account the temperature of reaction because the lower the temperature, the easier the reaction is to control. Figures 5 and 6 show that LDPE has a reduced domain of correct functioning with regard to that of HDPE. It comes essentially from Its strong reactivity to ozone. After having defined the limits of this installation, we studied the effect of different parameters on the quality of the ozonization such as crystallinity and granulometry of polymers. lime of ozonization and effect of the load of reactor.
Crystallinity of Polymers
The ozonization of powdered PE by air/ozone mixture is a gas/solid reaction. The solubility and the diffusivity of ozone into PE are important factors. Of all gases, ozone is the most soluble. However, the permeability of polyolefin to ozone depends essentially upon its crystallinity. Bodrero (29) uses experimental relations that consider that the amorphous phasis is permeable and that the crystalline phasis is not permeable. Then, we compared the peroxidation of HDPE and LDPE powders with similar granulometric size under the same experimental conditions (Table 3). The content of peroxide species on HDPE is superior to that measured on LDPE, and this difference increases with time of ozonization. However, it is known that HDPE is less reactive than LDPE since it is more crystalline, and it possesses a linear structure, which ensures it a greater chemical inertia. Moreover, HDPE possesses relatively few tertiary carbons with regard to LDPE which is branched (Table 1). These results could arise from the porosit y of the surface of particles, which is not known, and the reactivity of PE to ozone, which depends upon the catalytic residues (case of HDPE used) and the stabilizing agent added.
Granulome try of Polymers
The penetration of ozone inside particles is not immediate since the solubility and the diffusivity of ozone in PE are not infinite parameters. If the exchange surface of the grain is more important, the peroxidation sites are more accessible. The peroxidation can vary with the size of particles, the factor of form and the ratio surface/volume. HDPE powder with large granulometric dispersion (100-800) has been ozonized and then sifted through sifters of different mesh size. The size of particles has little effect on the peroxidation of HDPE (200-800). Inversely, thin particles (100-200) coming from a fluidized bed are more peroxided: 6.1 x [10.sup.-5] eq/g Instead of 5.2 x [10.sup.-5] eq/g for greater diameters. To obtain a very homogeneous ozonization, it is preferable to separate thin particles. These being sensitive to the electrostatic phenomenon, they tend to stick on the internal faces of the reactor.
The quality of the fluidization plays a preventive role against the risks of reaction acceleration. However, it determines a major role in the activation of the polymer. So, the same powdered polymer is ozonized in two reactors, which possess columns of different diameters and differ by their type of distributor. Table 4 gives the activation rates of ozonized LDPE as a function of experimental conditions used and for a constant speed of fluidization. The reactor of 40-mm diameter is a reactor with a double envelope, inside of which flows a thermostated oil. The difference in temperature between the oil and the fluidized layer is 7[degrees]C to 8[degrees]C under the experimental conditions used. Runs 3 and 5 were achieved using the same experimental conditions except for the quality of fluidization. They show that the quality of fluidization slightly affects the activation of polymer since the rate of activation varies from 3.50 to 3.75 x [10.sup.-5] eq/g. This results from the relatively short time of ozoniza tion (45 min) and the moderate temperature (38[degrees]C). The rate of peroxidation is also little influenced by the fluidization (pistoning or bubbling). Runs 2 and 4 show that the quality of fluidization has an effect on the rate of activation for important ozonization times (120 min) and high peroxidation rates (Table 4). Under the experimental conditions, the rate of activation increases from 5.30 to 6.85 x [10.sup.-5] eq/g. This result comes from the experimental parameters used ([T.sub.R] = 44[degrees]C instead of 40. [Yo.sub.3] = 23 g/[m.sup.3] instead of 15). The peroxidation rate of polyethylene varies with the quality of fluidization because the ozonization conditions used are drastic (important ozonization time, high reaction temperature and important peroxide rate).
Time of Ozonization
Figure 7 shows that the time of ozonization strongly influences the peroxidation rate and that the activation of polymer is a function of [square root of (t)] for temperatures of reaction about 40-45[degrees]C. These results are in agreement with those obtained by Verney and Michel (5). To explain these results, two hypotheses are given:
* the active sites rarefy during the time.
* the decomposing peroxides and particularly, hydroperoxides, resulting from the temperature of the reactor and the radical environment could react with these peroxides.
This second hypothesis can also be applied to the results of Sarraf (30), which noted that the proportion of hydroperoxides with regard to total peroxide species is 60% after an ozonization at 20[degrees]C, whereas this proportion is 35% after an ozonizalion at 45[degrees]C. Consequently, the hydroperoxide groups would be more sensitive to this decomposition, which increases with the temperature.
Effect of the Load of Reactor
The amount of polymer added in the reactor can influence its activation. Consequently, when the load increases, two parameters can decrease the peroxide rate:
* the ozone concentration in the high part of the fluidized bed can decrease,
* the quality of the fluidization can deteriorate.
We noted that the increase in the amount of PE (+ 50%) does not affect the activation of polymer since the quality of fluidization Is not really modified. It should be noted that the peroxide rate can be more sensitive to the increase of the mass when it leads to less good fluidization.
Optimization of Experimental Conditions
To optimize the ozonization reaction of HDPE 200-600, we used a well-known experimental method (31). The effect of the temperature [X.sub.1] (lower limit = 25[degrees]C and upper limit = 45[degrees]C), the time [X.sub.2] (lower limit = 20 min and upper limit = 60 min) and the throughput of air/ozone gas flow [X.sub.3] (lower limit = 0.8 [m.sup.3]/h and upper limit = 1.2 [m.sup.3]/h has been studied. Note that the production of ozone is equal to 16 g/h. With these parameters, we have to operate [2.sup.3] reactions. We considered only the lower and upper limit for each parameter. The matrix of the experiment is summed up in Table 5. The theoretical answer [Y.sub.th] defined as a function of experimental conditions can be expressed by Eq 4.
[Y.sub.th] = [b.sub.0] + [b.sub.1][X.sub.1] + [b.sub.2][X.sub.2] + [b.sub.3][X.sub.3] + [b.sub.12][X.sub.1][X.sub.2] + [b.sub.13][X.sub.1][X.sub.3] + [b.sub.23][X.sub.2][X.sub.3] + [b.sub.123][X.sub.1][X.sub.2][X.sub.3] (4)
where [b.sub.0] = [[SIGMA].sub.i][Y.sub.iexp]/[2.sup.3]; [b.sub.j] = [[SIGMA].sub.i][Y.sub.iexp] * [X.sub.j]/[2.sup.3]; [b.sub.jk] = [[SIGMA].sub.i][Y.sub.iexp] * [X.sub.j][X.sub.k]/[2.sup.3]
[b.sub.123] = [[SIGMA].sub.i][Y.sub.iexp] * [X.sub.1][X.sub.2][X.sub.3]/[2.sup.3] and [Y.sub.exp]: peroxide rate determined by iodometric method (Too).
Table 6 shows that the temperature has a greater effect than the time or the throughput of air/ozone gas flow on the peroxide rate since [b.sub.1] > [b.sub.2] > [absolute value of [b.sub.3]]. We note that the influence of throughput is weak and acts in the opposite direction since [b.sub.3] < 0. When the throughput increases, the ozone concentration decreases and then the peroxide rate decreases. Yet this result must be considered with caution because the reduction of throughput leads to a decrease in the quality of fluidization, notably at the end of the ozonization when the particles of PE tended to agglomerate. Coefficients [b.sub.jk] being negligible except for [b.sub.12], we can express the theoretical response [Y.sub.th] by Eq 5.
[Y.sub.th] * [10.sup.5] = 4.063 + 0.938 [X.sub.1]
+ 0.663 [X.sub.2] - 0.14 [X.sub.3] + 0.288 [X.sub.1][X.sub.2] (5)
The ozonization corresponding to [X.sub.1] = [X.sub.2] = [X.sub.3] = 0 (35[degrees]C, 40 min, 1 [m.sup.3]/h) gives a result close to the theoretical result ([Y.sub.exp] * [10.sup.5] = 3.8 and [Y.sub.th] * [10.sup.5] = 4.06).
Characterization of Ozonized Polymers
Experimental parameters being defined, we tried to characterize ozonized HDPE and LDPE. So, infrared spectroscopy of thin films must be performed. If the optical density of a particular species and the thickness of the film are known, the double bond concentration C of that species can be calculated from Eq 6.
C = A / [epsilon]*[rho]*l (6)
where A is the observed absorbance, [epsilon] is the extinction coefficient (32) of the species (L/mol cm), [rho] is the density (g/L) and l is the thickness of the film (cm).
The sites preferentially attacked by ozone are double bonds such as vinyl, vinylene and vinylidene and other groups such as tertiary carbons. Table 7 gives the concentrations in functional groups expressed per 1000 carbons atoms before and after ozonization of HDPE:
* vinyl double bonds located at the end of chains are numerous ([[-CH = [CH.sub.2]].sub.0] = 5.85 X [10.sup.-5] mol/g) and the most reactive to ozone (- 4.75 X [10.sup.-5] mol/g).
* vinylidene double bonds are hanging bonds which are fewer than vinyl groups,
Their sum is equal to total number of groups (= [CH.sub.2]).
* few trans-vinylene double bonds [[-CH = CH-].sub.0] = 0.41 x [10.sup.-5] mol/g, little reactive to ozone (- 0.09 X [10.sup.5] mol/g),
* inversely, (-CH = CH - CH = CH-) are more reactive to ozone (- 0.60 X [10.sup.-5] mol/g) but are also few (0.70 x [10.sup.-5] mol/g),
Their sum is equal to the number of intramolecular double bonds.
Methyl groups could give some information about the amount of branching. However, the characteristic bands of these groups are masked by other more intense bands.
These results show that double bonds (particularly vinyl groups) of HDPE have been attacked strongly by ozone. Indeed, 80% of the double bonds attacked (4.92 X [10.sup.-5] mol/g) are (= [CH.sub.2]) groups. The oxidation of PE can be followed by observing the increase in intensity of the carbonyl groups at the appropriate wavelength (1720 [cm.sup.-1]). Table 7 shows that the variation in carbonyl groups (12.27 X [10.sup.-5] mol/g) is about double the variation in the double bonds. The literature shows that ozone attacks double bonds by formation of unstable ozonides, which lead to the formation of carbonyl groups, preferentially acid groups (33). However, we cannot attribute the formation of the carbonyl groups only to the disappearance of double bonds, since the peroxide and hydroperoxide groups also generate carbonyl groups (14). Consequently, infrared spectroscopy can quickly give an Idea of the peroxidation, but it cannot be used to quantify the peroxide rate contained in ozonized PE. This analysis permits verification that the double bonds are preferential sites of oxidation by ozone.
In the second step, we determined the weight_and number-average molecular masses ([M.sub.w] and [M.sub.n]) of samples by SEC. Table 8 shows that [M.sub.w] of HDPE, is divided by a factor two, whereas [M.sub.n] remains almost constant. Consequently, the polymolecularity index decreases strongly from 12.3 to 7, indicating a reduction in the breadth of the molecular distribution. These results show that the sites attacked by ozone (tertiary carbons and trans-vinylene double bonds) are dispersed along the chain and that their number is more or less proportional to the length of the chain. Table 8 shows that the ozonization attacks the weak and high masses of LDPE since [M.sub.w] and [M.sub.n] decrease, whereas the polymolecularity index tends to increase. The ozonization reacts on the tertiary carbons, generating intramolecular dialkylperoxides. The results presented in Table 9 are obtained from Eqs 7 to 11.
[T.sub.OO] = [T.sub.ROOR] + [T.sub.ROOH],
oxidation rate determined by iodometric method (7)
[n.sub.ROOR] = [Mn.sub.h0]/[Mn.sub.h17] - 1, number of peroxide by chain
(peroxides and hydroperoxides) (8)
[DP.sub.n] = [M.sub.n] / 28 (9)
[n.sub.00] = [Too.sub.h17] * [Mn.sub.h0],
number of peroxide groups by chain
([T.sub.ooh17] is measured by iodometric method) (10)
[n.sub.00]/1000C = [n.sub.oo] * 1000/[DPn.sub.h0] number of peroxide groups per 1000 carbon atoms (11)
The peroxide rate determined by the iodometric method for HDPE ([h.sub.17]) corresponds to an average of 0.5 peroxide function by chain of polymer, which corresponds to 25% of dialkylperoxides and 75% of hydroperoxides. The peroxide rate for LDPE ([b.sub.20]) corresponds to an average of 0.7 peroxide function by chain of polymer, which corresponds to 80% of dialkylperoxides and 20% of hydroperoxides. These results are in agreement with those of Sarraf (30), who obtained 67% of dialkylperoxides onto ozonized LDPE at 4400.
This study concerned the optimization and the explanation of the phenomena occurring during the ozonization of HDPE and LDPE in a fluidized bed, which required perfect control and identification of species formed during this treatment. The comparison of different methods of titration described in this work shows that the iodometric method and indirect titration by thiol are the most reliable and the most reproducible methods. Then, we proved that the reaction of ozonization in a fluidized bed is very exothermic and requires a system of thermal regulation to avoid the acceleration of the reaction and the degradation of polymers. From a thermally regulated reactor, we evaluated the essential parameters controlling the reaction of ozonization. Then, we showed that HDPE is more sensitive to ozone than LDPE and that the size of particles of polymers slightly influences the peroxide rate. Experimental parameters such as time and temperature of ozonization strongly influence the peroxide rate. This study also conce rned the characterization of the structure of oxidized polymers. The double bonds and the tertiary carbons are attacked strongly by ozone, which leads to the modification of [M.sub.w] and [M.sub.n]. Then, we showed that ozonized HDPE leads essentially to the formation of hydroperoxide groups, whereas ozonized LDPE, rather, involves the formation of peroxide groups.
(1.) UMR 5076 - Laboratoire de Chimie Macromoleculaire Ecole Nationale Superieure de Chimie de Montpellier 8, rue de l'Ecole Normale 34296 Montpellier Cedex 5, France
(2.) C.E.RE.MA.P Route des Salins, B.P. 118 - 34140 Meze, France
(*.) Corresponding author.
(1.) D. M. Brewis and D. Briggs, Polymer, 22, 7 (1981).
(2.) Y. Miwa, K. Yamamoto, M. Sakaguchi, and S. Shimada, Macromolecules, 32, 8234 (1999).
(3.) A. Bradley, Chemtech, August 1973, p. 507.
(4.) S. Yamakawa and F. Yamamoto, J. Appl. Polym. Sci., 25, 41 (1980).
(5.) V. Verney and A. Michel, Polym. Process Eng., 4, 321 (1986).
(6.) C. Polenick, H. Widdecke, S. Bodrero, A. Revillon, and A. Guyot, J. Polym. Sci., Polym. Chem., 28, 261 (1990).
(7.) B. Boutevin, Y. Pietrasanta, M. Taha, and T. Sarraf, Eur. Polym. J., 20, 875 (1984).
(8.) B. Boutevin, Y. Pietrasanta, M. Taha, and T. Sarraf, Eur. Polym. J., 20, 1131 (1984).
(9.) B. Boutevin, Y. Pietrasanta, and J. J. Robin, Eur. Polym. J., 27. 815 (1991).
(10.) B. Boutevin, J. J. Robin, and A. Serdani, Eur. Polym. J., 28, 1507 (1992).
(11.) T. Fargere, M. Abdennadher, M. Delmas, and B. Boutevin, J. Polym. Sci., Polym. Chem, 32, 1377 (1994).
(12.) T. Fargere, M. Abdennadher, M. Delmas, and B. Boutevin, Eur. Polym. J., 31, 489 (1995).
(13.) B. Boutevin, Y. Pletrasanta, J. J. Robin, and T. Poilet, Eur. Polym. J., 24, 953 (1988).
(14.) A. A. Kefeli, S. D. Razumovskii, and G. E. Zaikov, Vysokomol. Soed. A 13, 803 (1971).
(15.) N. Gaylord and M. K. Mishra, J. Polym. Sci. Polym. Lett. Ed., 21, 23 (1983).
(16.) M. Lambla. Comprehensive Polymer. Sci., First Supplement, p. 619, G. Allen, ed., Pergamon Press, New York (1982).
(17.) J. M. Lusinchi, B. Boutevin, N. Torres, and J. J. Robin, J. Appl. Polym. Sci., 79, 874 (2001).
(18.) Y. Morita, N. Suzuki, and H. Goto, U.S. Patent 4,839,432 (1989).
(19.) B. Boutevin, Y. Pietrasanta, J. J. Robin, and N. Torres, Makromol. Chem. Phys., 200, 142 (1998).
(20.) B. Boutevin, J. M. Lusinchi, Y. Pietrasanta, and J. J. Robin, Polym. Eng. Sci., 36, 879 (1996).
(21.) N. Torres, J. J. Robin, and B. Boutevin, J. Appl. Polym. Sci., submitted.
(22.) P. Tacke, H. Korber, J. Merten, and D. Nevray, U.S. Patent 4,739,017 (1988).
(23.) C. Brondino, doctoral thesis, University of Montpellier II (1996).
(24.) C. D. Wagner, R. H. Smith, and E. D. Peters, Ind. Eng. Chem. Anal (Anal Chem), 19, 976 (1947).
(25.) S. G. Gallo, H. K. Wiese, and J. F. Nelson, Ind. Eng. Chem., 40, 1277 (1948).
(26.) B. Ameduri, B. Boutevin, and M. Nouiri, J. Polym. Sci., Polym. Chem., 31, 2069 (1993).
(27.) A. J. Chirinos-Padron. P. H. Hernandez, E. Chavez, N. S. Allen, C. Vasiliou, and M. Depoortere, Eur. Polym. J., 23, 935 (1987).
(28.) J. Brandrup and E. H. Immergut, eds., Polymer Handbook 3rd Ed., Wiley, New York (1989).
(29.) S. Brugiere-Bodrero, doctoral thesis, University of Lyon I(1989).
(30.) T. Sarraf, doctoral thesis, University of Montpellier II (1987).
(31.) J. L. GuindoPlet, Caout. & Plast., 723, 79-86 (1993).
(32.) M. Dole, Advances in Radiaticn Chemistry, Vol. 4, 307, M. Burton and J. L. Magee, eds., Wiley, London (1974).
(33.) R. Criegee, Chimia, 22, 392 (1968).
[Figure 5 omitted]
[Figure 6 omitted]
Table 1 Characteristics of Polyethylenes (28). [M.sub.w] [10.sup.-3] Density Polymers (g/mol) (kg/[m.sup.3] -[CH.sub.3] HDPE Ziegler 50-60 960 3.6 HDPE Philips 50-60 965 3.1 HDPE Philips (gas process) 350 954 2.8 LDPE -- 918-928 20-30 Groups per 1000 carbon atoms Polymers -[CH.sub.2][CH.sub.3] -CH=[CH.sub.2] HDPE Ziegler 0.5 0.09 HDPE Philips 0.8 1.58 HDPE Philips (gas process) <0.2 0.65 LDPE 6-9 0.08-0.25 Groups per 1000 carbon atoms Polymers -CH=CH- =C=[CH.sub.2] HDPE Ziegler <0.02 0.06 HDPE Philips <0.02 0.08 HDPE Philips (gas process) <0.02 0.04 LDPE <0.02-0.06 0.17-0.33 Table 2 HDPE Unsaturations Determined by lodometric Method (25, 27. Type of PE [>C=C<]/mol % >C=C</1000C [10.sup.5].[>C=C<]/[g.sub.PE] PEhd Ziegler 0.385 1.9 13.8 PEhd Philips 0.481 2.4 17.2 Table 3 Effect of the Nature of PE on the Peroxidation for Powders With Similar Size ([m.sub.PE 200-600] = 50g). T [T.sub.R] PE Crystallinity (a) (min) ([degrees]C) Virgin HDPE 200-600 0.63 -- -- Ozonized HDPE 200-600 0.63 60 25 Ozonized HDPE 200-600 0.63 60 35 Ozonized HDPE 200-600 0.63 45 45 Ozonized (d) HDPE 200-600 0.63 120 44 Virgin LDPE 200-600 0.44 -- -- Ozonized LDPE 200-600 0.44 60 35 Ozonized LDPE 200-600 0.44 45 45 Ozonized (d) LDPE 200-600 0.44 120 44 [Q.sub.v] [Y.sub.03] PE ([m.sup.3]/h) (g/[m.sup.3]) Virgin HDPE 200-600 -- -- Ozonized HDPE 200-600 1.2 13.3 Ozonized HDPE 200-600 1.2 15 Ozonized HDPE 200-600 1.2 15 Ozonized (d) HDPE 200-600 0.77 23 Virgin LDPE 200-600 -- -- Ozonized LDPE 200-600 1.2 15 Ozonized LDPE 200-600 1.2 15 Ozonized (d) LDPE 200-600 0.77 23 [T.sub.[infinity]] [T.sub.o] [10.sup.5] [10.sup.5] PE (eq/g) (b) (eq/g). (c) Virgin HDPE 200-600 2.90 [+ or -] 0.05 0 Ozonized HDPE 200-600 9.5 3.30 [+ or -] 0.1 Ozonized HDPE 200-600 12.3 4.7 [+ or -] 0.1 Ozonized HDPE 200-600 14.6 5.85 [+ or -] 0.1 Ozonized (d) HDPE 200-600 17.4 7.25 [+ or -] 0.1 Virgin LDPE 200-600 0.30 [+ or -] 0.05 0 Ozonized LDPE 200-600 6.9 3.30 [+ or -] 0.1 Ozonized LDPE 200-600 10.1 4.90 [+ or -] 0.1 Ozonized (d) LDPE 200-600 10.9 5.30 [+ or -] 0.1 (a):Data given by DSM. (b):[T.sub.o] = oxidation rate determined by lodometric method. (c):[T.sub.[infinity]] = peroxide rate ([T.sub.[infinity]] = ([T.sub.o] - [T.sub.o] virgin)/2). (d):Ozonization in the reactor of diameter = 40 mm. Table 4 Effect of the Fluidization Quality on the Ozonization Efficiency ([m.sub.LDPE 200-600] = 50g). Quality of t [T.sub.oil] [T.sub.R] [n.sup.o] Fluidization (min) ([degrees]C) ([degrees]C) 1 -- -- -- -- 2 (a) pistoning 120 52 44 3 (a) pistoning 45 45 38 4 (b) bubbling 120 -- 40 5 (b) bubbling 45 -- 38 [Q.sub.v] U [Y.sub.03] [Pr.sub.03] [n.sup.o] ([m.sup.3]/h) (cm/s) (g/[m.sup.3]) (g/h) 1 -- -- -- -- 2 (a) 0.77 17.0 23 16 3 (a) 0.77 17.0 15 11.5 4 (b) 1.2 17.0 15 18 5 (b) 1.2 17.0 15 18 [T.sub.o] [10.sup.-5] [T.sub.[infinity]] [10.sup.-5] [n.sup.o] (eq/g) (eq/g) 1 3.30 [+ or -] 0.05 0 2 (a) 10.90 5.30 [+ or -] 0.10 3 (a) 7.30 3.50 [+ or -] 0.10 4 (b) 14.00 6.85 [+ or -] 0.10 5 (b) 7.80 3.75 [+ or -] 0.10 Type of reactor: (a)simple reactor with double envelope; (b)fluidized and thermostated bed. Table 5 Matrix of Experiment([m.sub.HDPE 200-600] = 50g; [Pr.sub.O3] = 16g/h). [X.sub.1] [X.sub.1] n[degrees] [X.sub.1] [X.sub.2] [X.sub.3] [X.sub.2] [X.sub.3] [h.sub.1] +1 +1 +1 +1 +1 [h.sub.2] -1 +1 +1 -1 -1 [h.sub.3] +1 -1 +1 -1 +1 [h.sub.4] -1 -1 +1 +1 -1 [h.sub.5] +1 +1 -1 +1 -1 [h.sub.6] -1 +1 -1 -1 +1 [h.sub.7] +1 -1 -1 -1 -1 [h.sub.8] -1 -1 -1 +1 +1 [h.sub.9] & 0 0 0 0 0 [h.sub.10] [X.sub.1] [X.sub.2] [X.sub.2] [Y.sub.exp] n[degrees] [X.sub.3] [X.sub.3] [10.sup.5] [h.sub.1] +1 +1 5.65 [+ or -] 0.1 [h.sub.2] +1 -1 3.30 [+ or -] 0.1 [h.sub.3] -1 -1 3.85 [+ or -] 0.1 [h.sub.4] -1 +1 2.80 [+ or -] 0.1 [h.sub.5] -1 -1 6.15 [+ or -] 0.1 [h.sub.6] -1 +1 3.65 [+ or -] 0.1 [h.sub.7] +1 +1 4.20 [+ or -] 0.1 [h.sub.8] +1 -1 2.65 [+ or -] 0.1 [h.sub.9] & 0 0 3.80 [+ or -] 0.1 [h.sub.10] Table 6 [B.sub.j] Coefficients. [b.sub.0] 4.063 [b.sub.1] 0.938 [b.sub.2] 0.663 [b.sub.3] -0.14 [b.sub.12] 0.288 [b.sub.13] -0.06 [b.sub.23] -0.09 [b.sub.123] 0.038 Table 7 Concentration in Functional Groups Determined by Infrared Spectrometry Before and After Ozonization of HDPE 200-600 With [T.sub.[infinity]] = 5.7 x [10.sup.-5] eq/g. v Groups ([cm.sup.-1]) [R.sub.1][R.sub.2]C = [CH.sub.2]:vinylidene 890 -CH = [CH.sub.2]:vinyl 910 [SIGMA] (= [CH.sub.2]) -CH = CH-:trans-vinylene 965 -CH = CH - CH = CH- 990 [SIGMA] (CH = CH) [SIGMA] (=) -[CH.sub.3] -- [R.sub.1][R.sub.2]C = O 1720 [epsilon] (32) Groups (L/mol cm) [R.sub.1][R.sub.2]C = [CH.sub.2]:vinylidene 129 -CH = [CH.sub.2]:vinyl 122 [SIGMA] (= [CH.sub.2]) -CH = CH-:trans-vinylene 169 -CH = CH - CH = CH- 360 [SIGMA] (CH = CH) [SIGMA] (=) -[CH.sub.3] -- [R.sub.1][R.sub.2]C = O 188 C ([10.sup.5] mol/g) Groups HDPE [R.sub.1][R.sub.2]C = [CH.sub.2]:vinylidene 0.45 -CH = [CH.sub.2]:vinyl 5.85 [SIGMA] (= [CH.sub.2]) 6.30 -CH = CH-:trans-vinylene 0.41 -CH = CH - CH = CH- 0.70 [SIGMA] (CH = CH) 1.81 [SIGMA] (=) 8.11 -[CH.sub.3] -- [R.sub.1][R.sub.2]C = O 2.39 C ([10.sup.5] mol/g) ozonized Groups [HDPE.sup.a] [R.sub.1][R.sub.2]C = [CH.sub.2]:vinylidene 0.28 -CH = [CH.sub.2]:vinyl 1.10 [SIGMA] (= [CH.sub.2]) 1.38 -CH = CH-:trans-vinylene 3.32 -CH = CH - CH = CH- 0.10 [SIGMA] (CH = CH) 0.52 [SIGMA] (=) 1.90 -[CH.sub.3] -- [R.sub.1][R.sub.2]C = O 14.67 Groups [DELTA] [R.sub.1][R.sub.2]C = [CH.sub.2]:vinylidene -0.17 -CH = [CH.sub.2]:vinyl -4.75 [SIGMA] (= [CH.sub.2]) -4.92 -CH = CH-:trans-vinylene -0.09 -CH = CH - CH = CH- -0.60 [SIGMA] (CH = CH) -1.29 [SIGMA] (=) -6.21 -[CH.sub.3] -- [R.sub.1][R.sub.2]C = O 12.27 (a.)(experimental conditions for ozonization: [m.sub.HDPE 200-600] = 50 g; t = 45 min; [T.sub.R] = 45[degrees]C; [Q.sub.v] = 1.2 [m.sup.3]/h; [Pr.sub.O3] = 18 g/[m.sup.3]; [Y.sub.O3] 15 g/[m.sup.3]) Table 8 Determination of Molar Masses by SEC. [T.sub.oo][10.sup.5] [M.sub.n] References (eq/g) (g/mol) Virgin HDPE [h.sub.0] 0 9000 Ozonized [HDPE.sup.a] [h.sub.17] (a) 5.7 8000 Virgin LDPE [b.sub.0] 0 11,700 Ozonized [LDPE.sup.b] [b.sub.17] (b) 6.75 7900 [M.sub.w] Polymolecularity (g/mol) Index Virgin HDPE 111,00 12.3 Ozonized [HDPE.sup.a] 55,700 7 Virgin LDPE 54,500 4.6 Ozonized [LDPE.sup.b] 42,800 5.4 Experimental conditions for ozonization: (a)([m.sub.HDPE 200-600] = 50 g; t = 45 min; [T.sub.R] = 45[degrees]C; [Q.sub.V] = 1.2 [m.sup.3]/h; [Pr.sub.O3] = 18 g/[m.sup.3]; [Y.sub.O3] = 15 g/[m.sup.3]) (b)([m.sub.LDPE 200-600] = 50 g; t = 60 min; [T.sub.R] = 40[degrees]C; [Q.sub.V] = 1.2 [m.sup.3]/h; [Pr.sub.O3] = 18 g/[m.sup.3]; [Y.sub.O3] = 15 g/[m.sup.3]) Table 9 Number of Peroxides and Hydroperoxides Created by the Ozonization. [T.sub.[infinity] [M.sub.n] [10.sup.5] (eq/g) (g/mol) [DP.sub.n] HDPE 0.0 9010 643 ozonized HDPE (a) 5.70 8000 571 LDPE 0.0 10,370 741 ozonized LDPE (b) 6.75 6630 473 [N.sub.[infinity]] [n.sub.[infinity]] [n.sub.roor] /chain /1000C /chain HDPE -- -- -- ozonized HDPE (a) 0.514 0.800 0.127 LDPE -- -- -- ozonized LDPE (b) 0.700 0.945 0.565 [n.sub.roor] % % /1000C ROOR ROOH HDPE -- -- -- ozonized HDPE (a) 0.198 24.7 75.3 LDPE -- -- -- ozonized LDPE (b) 0.762 80.7 19.3 Experimental conditions for ozonization: (a)([m.sub.HDPE 200-600] = 50 g; t = 45 min; [T.sub.R] = 45[degrees]C; [Q.sub.V] = 1.2 [m.sup.3]/h; [Pr.sub.O3] = 18 g/[m.sup.3]; [Y.sub.O3] = 15 g/[m.sup.3]) (b)([m.sub.LDPE 200-600] = 50 g; t = 120 min; [T.sub.R] = 40[degrees]C; [Q.sub.V] = 1.2 [m.sup.3]/h; [Pr.sub.O3] = 18 g/[m.sup.3]; [Y.sub.O3] = 15 g/[m.sup.3])
|Printer friendly Cite/link Email Feedback|
|Author:||Boutevin, B.; Robin, J.J.; Torres, N.; Casteil, J.|
|Publication:||Polymer Engineering and Science|
|Article Type:||Statistical Data Included|
|Date:||Jan 1, 2002|
|Previous Article:||A new process for crosslinking thermoplastic polyurethanes using rubber industry techniques.|
|Next Article:||Percolation phenomena in polymers containing dispersed iron.|