Influence of transition metal-doped titanium(IV) dioxide on the photodegradation of polyethylene

Polymer Degradation and Stability 91 (2006) 3010e3019www.elsevier.com/locate/polydegstab

Influence of transition metal-doped titanium(IV) dioxideon the photodegradation of polystyrene

Terence J. Kemp*, Robin A. McIntyre

Department of Chemistry, University of Warwick, Coventry CV4 7AL, UK

Received 30 March 2006; received in revised form 24 July 2006; accepted 9 August 2006

Available online 10 October 2006

Abstract

The effects of adding dispersed powders of various forms of titanium(IV) dioxide on the photodegradation of polystyrene have been exam-ined using FT-IR spectroscopy from the following points of view: effect of crystal form, concentration of pigment, transition metal ion, dopantconcentration, calcination temperature of pigment, and pigment coating.

The rate of photodegradation of polystyrene is reduced by adding certain grades of TiO2 such as coated TiO2 particles or TiO2 doped withsmall percentages of Cr or Mn ions. The rate is increased on adding TiO2 doped with V and especially Mo or W ions. The anatase form of TiO2 ismore photoactive than the rutile form, as is the effect of increasing the calcination temperature of the pigment. The concentration dependences ofthe degradation rates are complex, but can be directly related to the percentage of anatase achieved after calcination. Even the most aggressive ofthe metal-doped pigments are less photoactive than a Degussa P25 material, containing rutile and anatase.� 2006 Elsevier Ltd. All rights reserved.

Keywords: Anatase; Dopant; Photodegradation; Polystyrene; Titanium dioxide; Transition metal

1. Introduction

The incorporation of titanium dioxide into bulk polymers isconducted on a very large scale in order to achieve a white col-oration and improved material properties. Such incorporationis not without its drawbacks, however, and efforts have beenmade to alleviate these by modifying the TiO2 [1].

In an earlier paper [2] we described the effects of modifyingtitanium dioxide by various means, in particular doping withlow percentages of a range of transition metal ions, on the ratesof (a) carbonyl group production in, and (b) chloride ion loss by,poly(vinyl chloride) as film or particles. The photoactivity ofdoped TiO2 was found to be a complex function of dopant con-centration, the energy levels of the dopants in the TiO2 lattice,and their d-electron configuration and local distribution [2].Photoactivity was also linked to such factors as the crystaltype of TiO2, its particle-size distribution and surface area. In

* Corresponding author. Tel.: þ44 2476 523235; fax: þ44 2476 524112.

E-mail address: [email protected] (T.J. Kemp).

0141-3910/$ - see front matter � 2006 Elsevier Ltd. All rights reserved.

doi:10.1016/j.polymdegradstab.2006.08.005

the present paper we report an analogous study of the effect ofincorporating various forms of TiO2, doped and undoped, intopolystyrene, upon its photodegradation; this study was aimedat understanding the result of modifying the pigment as muchas that of the photodegradation of the polystyrene. The mecha-nism of weathering of polystyrene was the subject of a detailedsurvey by Lemaire et al. [3], who summarise the work of Mail-hot and Gardette [4,5] as well as that of earlier authors [6].Rabek [7] and Rosik [8] have also reviewed the photodegrada-tion of polystyrene and its analogues. Lemaire et al. [3] focusedon the role of irradiation wavelength, differentiating between‘long’ (l> 300 nm) and ‘short’ (l¼ 254 nm) wavelengths.The macroradical (1) has been identified as an intermediate byEPR spectroscopy [9];

2 2~ CH C Ph CH ~•

− −

1

this reacts with oxygen to form the corresponding peroxylradical (2) and hydroperoxide (3) which absorbs at 3450 cm�1.

mailto:[email protected]

http://www.elsevier.com/locate/polydegstab

3011T.J. Kemp, R.A. McIntyre / Polymer Degradation and Stability 91 (2006) 3010e3019

2 2~ CH CPh CH ~− −

2 3

OOH

2 2~ CH - CPh - CH ~

2O •

Hydroperoxide (3) undergoes photolysis or thermolysis toyield the alkoxyl radical (4), which can either abstract a hydro-gen atom from a neighbouring eCHPhe moiety to yield thehydroxy compound (5) (nabs 3450, 3540 cm�1) or undergoscission in two ways:

4

O•

2 2~ CH CPh CH ~− −

4 5

2 2~ CH C CH ~− − Ph+ •

2 2~ CH CPh CH ~− − 11725 cm−

2~ CH C Ph− −

11690 cm−

2 CH CHPh+ • − ~

6

O

O

OH

2 2~ CH CPh CH ~− −

O •

Further reactions of the macroradical (6) take place, anda complete scheme is provided by Lemaire et al. [3].

Recent papers deal with polystyrene containing various per-centages of TiO2 acting as a photocatalyst [10,11]. Here thekey intermediate, formed following excitation of the TiO2 par-ticle, is the �OH radical. This attacks the polystyrene by hydro-gen-atom abstraction at two sites:

~ 2 2CH C PhCH•

~ 2 H O+

OH• + ~ 2 2CH CHPh CH− ~

~ 2C HCHPhCH•

~ 2 H O+

The main carbonyl photo-product absorbed at 1724 cm�1

but no assignment was made [10].The present paper examines the effects of modifying the

dispersed TiO2 pigment on the susceptibility of polystyreneto UV-induced degradation.

2. Experimental

2.1. Materials

Chemicals were commercial materials of the highest avail-able purity. Polystyrene of average Mn 230 000, Mw ca.

140 000 was obtained from Aldrich. Titania pigments wereas follows: P25 was from Degussa AG, K2220 from Kronos,RM pigments were made in-house, the number code refers tothe calcination temperature in degrees Celcius, DT pigmentswere supplied by Millennium Chemicals and refer toW- and Mo-doped materials, respectively. PC500 is a Millen-nium Chemicals’ sample of anatase.

2.2. Preparation of doped TiO2

Methods of preparation of doped TiO2, both by absorptionand coprecipitation, are given in detail in Ref [2]. All samplesof titania that were modified in-house had sub-micron particlesizes.

2.3. Preparation of polystyrene films

The same method was used as for PVC films [2] except thatthe HPLC grade toluene was used as the solvent for polysty-rene. This approach was considered appropriate in view ofthe limited quantities of specially modified pigment thatwere available. The films were of nominally 150 mm thicknessas confirmed (�5%) by high-resolution optical microscopy[2]. We would draw attention to the fact that the results ob-tained using samples prepared in this way may not be strictlycomparable with those obtained using high shear melt mixing.

2.4. Instrumentation

The same range of instrumentation was used as for the studieson PVC [2]. The Q-Panel Weatherometer was equipped witheight fluorescent bulbs (300 W, UVB) and the light output cali-brated using ferrioxalate actinometry adapted for the Weather-ometer [2]. The output was measured as 2.98� 1017 quanta s�1.

3. Results and discussion

3.1. Non-pigmented polystyrene

Irradiation of a 150 mm film of polystyrene for 500 h gavea broad absorption in the carbonyl region with a maximum atca. 1743 cm�1.

3.2. Effect of loading with Degussa P25 pigment

Polystyrene films containing, respectively, 0, 4, 10 and20% w/w of TiO2 (Degusssa P25) were exposed in a QUVWeatherometer (UVB bulbs). The carbonyl index was moni-tored and plotted with respect to irradiation period (Fig. 1).

From Fig. 1 it is evident that upon increasing the loading ofTiO2, the rate of degradation increases owing to the fact thatDegussa P25 pigment is reported to possess high photoactivityand therefore will offer limited protection by filtering the lightentering the PS film. This result is in agreement with the studyof the effect of loading of Degussa P25 pigment upon the pho-todegradation of PVC film [2], the level of degradation beingless severe for the unpigmented polystyrene film for the sameirradiation time.

3012 T.J. Kemp, R.A. McIntyre / Polymer Degradation and Stability 91 (2006) 3010e3019

3.3. Effect of crystal type of undoped TiO2

Polystyrene films containing 4%, respectively, PC500 (ana-tase), Degussa P25 (70:30, anatase:rutile), RM600 (20:80,anatase:rutile) and RM1000 (rutile) were exposed in a QUVWeatherometer (UVB bulbs) and the carbonyl index was plot-ted with respect to irradiation time (Fig. 2).

Fig. 2 shows that, by increasing the rutile content of theTiO2 pigment, the rate of degradation decreases, i.e. on goingfrom PC500 to RM600 to RM1000 there is a marked reductionin the extent of degradation, a result which is similar to thatobserved in the degradation of PVC film using the same setof pigments [2]. It appears that the anatase-to-rutile ratio hasa significant effect upon the photocatalytic nature of the pig-ment, as anatase is generally regarded as the more photochem-ically active phase of titania.

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

0 100 200 300 400 500Time (h)

Ab

s C

O(1740)/A

bs C

H(2850)

PS (no titania) 4 % P2510 % P25 20 % P25

Effect of loading of Degussa P25 pigment upon

photodegradation of PS film

Fig. 1. Plot of carbonyl index with respect to time for different loadings of P25

in the PS matrix (150 mm thick films).

0

0.2

0.4

0.6

0.8

1

1.2

0 200 400 600 800Time (h)

Ab

s C

O(1714)/A

bs C

H(1460)

PE (no titania) PC500 RM600 RM1000 P25

Effect of crystal type of undoped titania upon the

photodegradation of PE film

Fig. 2. Chart depicting the effect of crystal type of TiO2 on extent of degrada-

tion of PS film with respect to irradiation period.

Degussa P25 pigment exhibits the most aggressive behav-iour towards the PS film, a result that was observed also inthe study of the photodegradation of PVC film [2]. Accordingto Gray et al. [10], this has been attributed to (i) the rutilephase acting as an antenna to extend the photoactivity to lon-ger wavelengths; (ii) the stabilization of charge separation byelectron transfer from rutile to anatase to slow recombination;(iii) the small size of the rutile crystallites facilitates this trans-fer, making catalytic ‘hotspots’ at the rutileeanatase interface.

3.4. Effect of TiO2 coating

Polystyrene films containing, respectively, Degussa P25(uncoated), RM1000 (uncoated rutile) and K2220 (coated ru-tile) were prepared at 4% loading and a thickness of150 mm. The samples were irradiated in the QUV Weatherom-eter (UVB bulbs) and the carbonyl formation was monitoredwith respect to irradiation period (Fig. 3).

From Fig. 3 it can be seen that the fastest rate of degrada-tion is produced by the uncoated P25 pigment whereas coatedTiO2 (K2220) has a major protective effect on the PS film,a finding which accords with the study of the photo-oxidativedegradation of PVC pigmented with coated and uncoated tita-nia pigments [2]. This result underlines the significance ofcoating the TiO2 with an impervious inorganic layer whichis thought to reduce the number of electrons and holes thatreach the surface, forming radicals, by encapsulation.

3.5. Vanadium-doped TiO2

3.5.1. Effect of calcination temperatureSamples of V(V)-doped TiO2 prepared by the coprecipita-

tion method, at a loading of 1% w/w, were calcined at two dif-ferent calcination temperatures, namely 873 and 1273 K.Polystyrene films were produced pigmented with 4% V-TiO2

and were exposed to UVB light in the QUV Weatherometer.The carbonyl indices were monitored with respect to irradia-tion time (Fig. 4).

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0 100 200 300 400 500Time (h)

Ab

s C

O(1740)/A

bs C

H(2850)

PS (no titania) 4% P25 K2220 (coated) RM1000

Effect of coating titania upon photodegradation

of PS film

Fig. 3. Graph illustrating the effect of coating TiO2 upon the photo-oxidative

degradation of PS film.


0

50

100

150

200

250

300

350

400

PS (no titania) RM600 RM1000 1% V (873K) 1% V (1273K)Sample

Tim

e taken

(h

) to

reach

C

.I. =

0.3

Effect of calcination temperature of V-doped titania upon


Fig. 4. Plot depicting the effect of calcination temperature of V-doped TiO2 (coprecipitation method) upon photo-oxidative degradation of PS film showing time

taken to reach carbonyl index of 0.3.

From Fig. 4 it is evident that none of the pigments showsignificant protection of the polymer film. The trend is oneof a slight decrease in degradation rate upon addition ofV-dopant ions compared with the corresponding undopedTiO2 at the same calcination temperature, as well as a furtherdecrease in degradation rate upon calcining at a higher temper-ature. This suggests that the crystal type of the TiO2 pigment,plus addition of V dopant ions, decreases the photoactivity andhence leads to lower degradation rates. The effects of loadinglevel of vanadium dopant and preparation method employed toproduce the pigment may also influence the degradation ratesalong with other factors, such as particle size, surface area andlocation of dopant ions.

Another interesting point to note is that the effect of addi-tion of titania pigments to the PS film is less marked, as com-pared to the degradation of PVC film with the same set ofpigments, due to the degradation rates of the various PS sam-ples being rather more similar [2]. In the degradation of PVC,an enhanced rate of formation of photo-oxidation products(carbonyl-containing species) is observed, which may be dueto the fact that the free Cl atoms produced in PVC can inducefurther photo-oxidation. In contrast, the degradation of PS filmappears to occur at a slower rate because it follows a different

degradation pathway, i.e. attack at the tertiary carbon positionin the polymer chain.

3.6. Effect of crystal type

Samples of V(V)-doped titania pigment were prepared viathe absorption method, using PC500 (undoped anatase) asa starting precursor, and calcined at 673, 873 and 1273 K pro-ducing V(V)-doped anatase, a V(V)-doped mixture of anataseand rutile and V(V)-doped rutile, respectively. Polystyrenesamples were prepared with 4% w/w V(V)-TiO2 or 4% TiO2

(undoped) and irradiated in the QUV Weatherometer (UVB

bulbs). The carbonyl index was monitored with respect to irra-diation period using FT-IR spectroscopy (Fig. 5).

From Fig. 5 it is noticeable that the rate of degradation de-creases according to the following sequence:

undoped anatase > V� doped anatase > undopedA=R

> V� dopedðA=RÞ > undoped rutile > V� doped rutile

> PSðno TiO2 addedÞ:

Undoped anatase and V-doped anatase show the greatestrates of degradation owing presumably to the fact that anatase

0

50

100

150

200

250

300

350

400

PS (notitania)

UndopedA/R

Undoped R 1% V A/R 1% V R 1% V A Undoped A

sample

Tim

e taken

(h

) to

reach

C

.I =

0.3

Effect of crystal type of V(V)-doped titania upon


Fig. 5. Graph displaying the effect of crystal phase of V(V)-doped TiO2 (absorption method) upon photo-oxidative degradation of PS film (150 mm) showing time

taken to reach carbonyl index of 0.3.


050

100150200250300350400450500

PS (no titania) RM600 1% Mn (873K) RM1000 1% Mn (1273K)sample

Tim

e taken

(h

) to

reach

C

.I. =

0.3

Effect of calcination temperature of Mn-doped titania upon


Fig. 6. Chart displaying the effect of calcination temperature of Mn-doped TiO2 (coprecipitation method) upon photo-oxidative degradation of PS film showing

time taken to reach carbonyl index of 0.3.

is more photoactive than rutile (see Section 3.3) whereasV-doped rutile shows the slowest rate of degradation for thepigmented polystyrene samples. All of the samples show noprotection towards the PS film. Again it is evident that addi-tion of V to the titania pigment reduces the degradation rate.This result is in agreement with the findings for the degrada-tion of PVC film [2].

3.7. Manganese-doped TiO2

3.7.1. Effect of calcination temperatureMn-doped titania pigments were prepared via the copreci-

pitation method (1% w/w loading level) and were calcined attwo different temperatures, namely 873 and 1273 K. Polysty-rene films were prepared pigmented with 4% Mn-doped titaniaor 4% undoped titania powders and at the usual thickness of150 mm. The films were then irradiated and the carbonyl de-velopment monitored as a function of time. (Fig. 6).

From Fig. 6 it is clear that the Mn-doped titania pigment,which has been calcined at 1273 K, is the only sample that of-fers significant protection towards the PS film. The trend

appears to be one of a decrease in the rate of degradationupon addition of Mn to the titania and by calcining at a highercalcination temperature. This is also in agreement with thestudy of effect of calcination temperature of V-doped TiO2

pigment upon the photodegradation of PS film (Section 3.5.1).Again it is interesting to note the closeness of the degrada-

tion rates, indicating that the effect of addition of pigment(both undoped and Mn-doped) is less distinct than in the studyof the degradation of PVC using the same set of pigments.

3.7.2. Effect of crystal typeSamples of Mn-doped TiO2 pigments were prepared by the

absorption method using PC500 as a precursor and at a loadinglevel of 1% w/w. The pigments were calcined at three differenttemperatures namely 673, 873 and 1273 K to yield Mn-dopedanatase, a Mn-doped mixture of anatase and rutile and Mn-doped rutile, respectively. The samples of both Mn-dopedTiO2 and their corresponding undoped TiO2 pigments wereembedded into PS film (150 mm thick) and irradiated. The car-bonyl index was plotted with respect to irradiation period(Fig. 7).

050

100150200250300350400450500

PS (no t

itania

)

PC500

MnPC50

0

RM600

1% M

n (87

3K)

RM1000

1%Mn (

1273

K)

sample

Tim

e taken

(h

) to

reach

C

.I. =

0.3

Effect of calcination temperature/crystal type upon


Fig. 7. Graph showing the effect of crystal type of Mn-doped TiO2 (absorption method) upon photo-oxidative degradation of PS film (150 mm) showing time taken

to reach carbonyl index of 0.3.


0

50

100

150

200

250

300

350

400

PS (no t

itania

)

RM600

0.1% M

o

0.3% M

o

0.5% M

o

0.7% M

o

0.9% M

o

1% M

o

sample

Tim

e taken

(h

) to

reach

C

.I. =

0.3

Effect of Mo loading level of Mo-doped titania (873K) upon


Fig. 8. Graph illustrating the effect of loading level of Mo in Mo-doped TiO2 (coprecipitation method, calcined at 873 K) upon photo-oxidative degradation of PS,

at different dopant levels showing time taken to reach carbonyl index of 0.3.

From Fig. 7 it is apparent that the rate of degradation de-creases according to the following sequence:

undoped anataseðPC500Þ > Mn� doped anatase

> undoped RM600 > Mn� doped TiO2ð873 KÞ> undoped rutileðRM1000Þ > PSðno titaniaÞ> Mn� doped rutileð1273 KÞ:

Upon increasing the calcination temperature, the crystaltype changes from anatase to the rutile phase and the rate ofdegradation decreases as a result. The effect is enhancedupon addition of manganese to the titania. This observationis in agreement with the corresponding study, using thesame pigments, of the effect of crystal type upon the photo-oxidative degradation of PVC film but in this case only theMn-doped rutile titania pigment offers protection towardsthe PS film.

3.8. Molybdenum-doped TiO2

3.8.1. Effect of loading and crystal typeMolybdenum(V)-doped TiO2 at loadings 0, 0.1, 0.3, 0.5,

0.7, 0.9 and 1% w/w were prepared via the coprecipitationmethod and were calcined at 873 K. Polystyrene films werethen produced with 4% Mo/TiO2 (thickness of 150 mm) andwere irradiated. The carbonyl development was plotted withrespect to irradiation time (Fig. 8).

From Fig. 8 it is evident that the 0.3 and 0.7% Mo-dopedTiO2 are the most aggressive pigments whereas the 0.5%Mo-doped TiO2 and the undoped RM600 show the slowestrates of degradation. The effect is shown more clearly inFig. 9 which displays the carbonyl index with respect tochange in loading level of Mo dopant for given irradiationperiods.

From Fig. 9 the trend appears to be one of an increase inreactivity from undoped to w0.3% Mo followed by a decreaseto 0.5% Mo. The trend then is an increase until w0.7% Mo

followed by a final decrease to 1% Mo. The trend has beenlinked [2] to change in the anatase-to-rutile ratio (as measuredby X-ray powder diffraction) as in the case for V- and Mn-doped TiO2 pigments and from the results for the study of ef-fect of Mo loading level in Mo-doped TiO2 pigment (calcinedat 873 K) upon the photodegradation of PVC film [2].

3.8.2. Effect of calcining at higher temperaturesMo-doped TiO2 pigments were also made at 0, 0.2, 0.4, 0.6

and 0.8% w/w, being prepared via the coprecipitation method,and were calcined at 1273 K. Polystyrene films were then pro-duced with 4% Mo/TiO2 at a thickness of 150 mm. The filmswere irradiated and the carbonyl development was monitoredwith respect to irradiation time (Fig. 10).

From Fig. 10 it is noticeable that the greatest rates of deg-radation are shown by 0.4 and 0.6% Mo-doped TiO2, respec-tively. All of the samples behave aggressively towards thePS film. Fig. 11 displays the effect of increasing the

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0 0.2 0.4 0.6 0.8 1% Mo loading level

Ab

s C

O(1740)/A

bs C

H(2850)

68 h

100 h

220 h

350 h

500 h

700 h

1000 h

Rate of degradation with respect to Mo loading

level in titania (873 K)

Fig. 9. Chart depicting the change in carbonyl index with respect to Mo load-

ing at a given irradiation time.


0

50

100

150

200

250

300

350

PS (no t

itania

)

RM1000

0.2% M

o

0.4% M

o

0.6% M

o

0.8% M

o

1% M

o

sample

Tim

e taken

(h

) to

reach

C.I. o

f 0.3

Effect of loading level of Mo-doped titania (1273K) upon


Fig. 10. Chart displaying the effect of dopant level in Mo-doped TiO2 (coprecipitation method, calcined at 1273 K) upon the photo-oxidative degradation of PS film

showing time taken to reach carbonyl index of 0.3.

molybdenum level in the titania upon the carbonyl index aftera given irradiation time.

This result is in agreement with the findings from thestudy of the photodegradation of PVC-containing Mo-dopedTiO2 (1273 K) [2] whereby the trend for both degradationtest reactions is similar; the maximum rate of degradationis shown by w0.4% Mo in both cases despite the great dif-ference in the detailed reaction mechanism for the twopolymers.

3.8.3. Comparison with Degussa P25 pigmentMo-doped titania pigments (one rutile, one anatase and one

a mixture of both) were dispersed into PS films at a loading of4% TiO2/PS (150 mm thick). Degussa P25 was also impreg-nated into PS film at the same loading and thickness for com-parison. The PS films were irradiated and the carbonyl indicesmonitored with respect to time (Fig. 12).

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0 0.2 0.4 0.6 0.8 1% Mo loading level

Ab

s C

O (1740)/A

bs C

H (2850)

68 h 164 h 265 h 425 h 750 h 930 h

Rate of degradation with respect to Mo loading

level (1273 K)

Fig. 11. Graph showing the change in rate of degradation upon increasing the

molybdenum level in titania (coprecipitation method, 1273 K).

From Fig. 12 it is evident that the greatest rate of deg-radation is experienced with the Degussa P25 pigment, asis the case for the comparison with Degussa P25 usingPVC as a test system [2]. All of the samples behave ag-gressively towards the PS film but the Mo-doped samplesare less aggressive than the standard Degussa P25. Thetrend is one of a decrease in aggressive behaviour for theMo-doped TiO2 pigments as the crystal type changesfrom anatase to rutile by increasing the calcination temper-ature during preparation.

3.9. Group(VI) metal-doped TiO2

3.9.1. Effect of doping with different group(VI) metal ionsSamples of doped TiO2 were made using Cr, Mo and W at

a loading of 1% w/w using the coprecipitation method fol-lowed by calcination at 873 K. The samples were placed

0

0.2

0.4

0.6

0.8

1

0 200 400 600 800 1000Time (h)

Ab

s C

O(1740)/A

bs C

H(2850)

PS (no titania)P25

Mo (mixture)Mo (rutile)

Mo (anatase)

A comparison of Mo-doped samples with Degussa

P25 using PS as a test sytem

Fig. 12. Plot showing the effect of Mo-doped titania samples, compared to De-

gussa P25 pigment, upon the photo-oxidative degradation of PS film.


0

50

100

150

200

250

300

350

PS (no titania) RM1000 1% Cr R 1% Mo R 1% W Rsample

Tim

e taken

to

reach

C

.I. =

0.3 Series1

Effect of group(VI) metal-doped titania upon photodegradation of PS film

Fig. 13. Graph illustrating the effect of doping TiO2 with a group(VI) metal ion upon the photo-oxidative degradation of PS film showing time taken to reach

carbonyl index of 0.3.

into PS films (150 mm thick) at a loading of 4% TiO2/PS. Thefilms were irradiated and carbonyl formation monitored withrespect to irradiation time elapsed (Fig. 13).

The results indicate that all the samples behave aggres-sively towards PS film apart from the Cr-doped rutile whichshows a protective effect. The trend is one of an increase inthe aggressiveness (increase in photoactivity) upon descendingthe group. This finding is in agreement to that obtained in thedegradation of PVC film using the same pigments [2].

Again it is important to note the apparent ‘‘bunching’’ ofthe degradation rates which has already been discussed in Sec-tion 3.5.1.

3.9.2. Effect of calcination temperature and crystal typeSamples of doped TiO2 were made as in Section 3.9.1 but

instead one set of doped powders was calcined at 673 and1273 K, producing doped anatase and rutile pigments. PSfilms were also prepared to the same thickness and loadingof doped titania. The carbonyl index for each sample was plot-ted with respect to irradiation time (Fig. 14).

From Fig. 14 it is clear that the greatest rates of degradationare shown by the Mo- and W-doped anatase pigments. Theslowest rate of degradation is shown by the Cr-doped rutilepigment. The trend upon changing crystal type from anataseto rutile is one of a reduction in rate of degradation, correlatingwith the results for the effect of group(VI) metal-doped TiO2

upon the photodegradation of PVC film [2]. Also there isa trend in increasing aggressive behaviour upon descendingthe group.

3.10. Doped anatase TiO2

3.10.1. Effect of doping anatase with V, Mn, Cr, Mo and WTiO2 (anatase) samples doped with Mn, V, Cr, Mo and W,

each with a loading of 1% w/w, were prepared via the absorp-tion method using PC500 (Millennium Chemicals) as a precur-sor and calcined at 673 K. PS films were then produced with4% TiO2 (thickness 150 mm) and irradiated. The carbonyl de-velopment was plotted with respect to irradiation time(Fig. 15).

0

50

100

150

200

250

300

350

PS (notitania)

1% Cr R 1% Mo R 1% W R 1% Cr A 1% Mo A 1% W A

sample

Tim

e tak

en

(h

) to

reach

C

.I. =

0.3

Effect of calcination temperature/crystal type upon photodegradation of PS film

Fig. 14. Chart displaying the effect of group(VI) metal-doped titania (calcined at 673 and 1273 K) upon photo-oxidative degradation of PS film showing time taken

to reach carbonyl index of 0.3.


0

50

100

150

200

250

300

350

PS (notitania)

PC500 VPC500 MnPC500 CrPC500 MoPC500 WPC500

sample

Tim

e taken

(h

) to

reach

C

.I. =

0.3

Series1

Effect of doping anatase with different TM ions upon photodegradation

of PS film

Fig. 15. Plot depicting the effect of change of transition metal for doped anatase upon photo-oxidative degradation of PS film showing time taken to reach carbonyl

index of 0.3.

From Fig. 15 it is noticeable that the Mo- and W-doped ti-tania pigments are behaving aggressively towards the PS filmas in Section 3.9.2. The Cr- and Mn-doped TiO2 pigments of-fer protection to the PS film. The degradation rates are reducedaccording to the following series:

TiO2=Mo > TiO2=W > undoped > TiO2=V > TiO2=Mn

> TiO2=Cr

This correlates reasonably well with the results obtained forthe photodegradation of PVC whereby the degradation ratesdecrease according to the following sequence [2]:

TiO2=Mo > TiO2=W > TiO2=V > undoped > TiO2=Mn

> TiO2=Cr

3.10.2. Comparison with Mo- and W-doped anatase pig-ments (supplied by Millennium Chemicals)

Mo- and W-doped titania pigments (anatase) were dis-persed into PS films at a loading of 4% TiO2/PS (150 mm

thick). DT52 and DT53 (W- and Mo-doped anatase TiO2,respectively, kindly supplied by Millennium Chemicals)were also impregnated into a PS film at the same loadingand thickness for comparison. The PS films were irradiatedand the carbonyl indices plotted with respect to time (Fig. 16).

From Fig. 16 it is apparent that the two industrially pro-duced pigments are more aggressive than the samplesproduced in-house. This could be due to several factors includ-ing: preparation method, calcination temperature, metal dop-ant loading level, surface area and particle-size distribution.

4. Conclusions

This paper describes the results from a study of the photo-oxidative degradation of PS film pigmented with undoped andtransition metal-doped TiO2 powders. Similar factors to thosestudied in the photo-oxidative degradation of PVC [2] havebeen investigated including: effect of coating undoped TiO2,effect of crystal type of both undoped and doped TiO2, effectof calcination temperature of doped TiO2, and effect of dopantmetal ion utilised.

0

50

100

150

200

250

300

350

PS (no titania) MoPC500 WPC500 DT52(10 wt% W)

DT53(10 wt% Mo)

sample

Tim

e taken

to

reach

C

.I. =

0.3

A comparison of Mo- and W-doped titania (produced in-house) with those

available commercially (Milennium Chemicals)

Fig. 16. Plot displaying the effect of Mo- and W-doped titania samples, compared to Mo- and W-doped pigment obtained from Millennium Chemicals, upon the

photo-oxidative degradation of PS film showing time taken to reach carbonyl index of 0.3.


The results for the undoped TiO2/PS system indicate thatthe anatase content of the pigment has a positive effect uponthe photocatalytic activity, and hence the relative amount ofdegradation the polymer undergoes. This observation correlatesreasonably well with the results for the undoped TiO2/PVC[2] system whereby increasing the amount of anatase presentin the sample significantly enhances the aggressiveness.Again the Degussa P25 pigment shows the greatest amount ofdegradation owing the fact to its exceptional photocatalyticactivity.

The effect of coating undoped titania again results in a pro-tective effect of the pigment towards PS film, an outcome thatwas experienced with the photodegradation of PVC film con-taining coated TiO2 pigment [2]. It is thought that encapsulat-ing the TiO2 particles in an impervious layer acts [2] to reducethe number of electrons/holes that can react at the surface.

The effect of doping the TiO2 with vanadium was investi-gated (but less extensively than for the study in our previouschapter using PVC [2]). The effects of calcination tempera-ture, and of the method of sample preparation, reveal that,upon increasing calcination temperature, the degradation ratedecreased, although not by sufficiently an extent to offer pro-tection to the PS film. Study of the crystal form of samplesmade via the absorption method has shown that V-doped rutileprovides the highest level of protection to the PS film, an im-provement in comparison to the undoped titania.

Doping titania pigments with manganese showed that sam-ples produced by both absorption and coprecipitation methodsand calcined at 1273 K provided protection to the PS film, at-tributable to the fact that the pigments consist entirely of therutile phase. The addition of manganese enhances the protec-tive nature of the pigment and hence reduces its photocatalyticactivity. These findings mirror those obtained for the Mn-doped TiO2/PVC system [2].

Doping titania with molybdenum enhances its photocata-lytic activity as measured by the photodegradation of PS film.The aggressive nature of the pigment is linked to its level of an-atase as is found for the photodegradation of PVC film using thesame Mo-doped TiO2 pigments. The optimum concentrationswere again found to be 0.3% Mo and 0.7% Mo when calcinedat 873 K and 0.4 and 0.6% Mo when calcined at 1273 K. Theseoptimum concentrations of Mo-doped TiO2 were tested along-side Degussa P25 but were found to be less aggressive.

The study of the effect of doping TiO2 with group(VI)metal ions revealed that W- and Mo-doped TiO2 pigmentsare particularly aggressive towards the PS film in both anataseand rutile crystalline forms. Cr-doped TiO2 offers protection tothe PS film, the protective effect increasing when the crystalphase is changed from anatase to rutile.

Doping anatase with Cr, Mn and V metal ions provides pro-tection to the PS film, whereas doping with Mo and W sensi-tises the PS film. The photo-oxidative degradation rates of PS

film pigmented with doped anatase decrease according to thefollowing sequence:

TiO2=Mo > TiO2=W > undopedTiO2 > PSðnotitaniaÞ> TiO2=V > TiO2=Mn > TiO2=Cr

This sequence parallels exactly that obtained for the photo-oxidative degradation of PVC film [2].

Mo- and W-doped TiO2 (anatase) pigments were also com-pared to samples of Mo and W-doped TiO2, obtained fromMillennium Chemicals, and were found to be less aggressive,owing possibly to the fact that the industrial samples containa higher concentration of dopant ions. Also the preparationconditions differ, resulting in dissimilar physical propertiessuch as particle size and surface area.

In conclusion, it is apparent that the results for the photode-gradation study of PS film correlate well with those obtainedin the degradation study of PVC film containing similar mod-ified titania pigments, although carbonyl growth generallyproceeds at a lower rate presumably due to differing degrada-tion pathways. In the case of PVC, the dehydrochlorinationreaction appears to result in significant formation of car-bonyl-containing photo-oxidation products.

Undoped rutile TiO2 is generally more protective than ana-tase, with coating improving this protective effect. Cr- andMn-doped TiO2 (rutile) reduce the degradation rate of PS film,whereas Mo- and W-doped TiO2 behave more aggressively.

Acknowledgements

We thank the EPSRC and Oxonica for financial supportthrough a Faraday Partnership. In particular, Drs Gareth Wake-field and Barry Park of Oxonica are thanked for valuable dis-cussion. Degussa, Kronos and Millennium Chemicals arethanked for providing research samples of TiO2.

References

[1] Kemp TJ, McIntyre RA. Prog React Kinet Mech 2001;26:337.

[2] Kemp TJ, McIntyre RA. Polym Degrad Stab 2005;91:165.

[3] Lemaire J, Gardette J-L, Mailhot B, Jouan X. In: Allen NS, Edge M,

Bellobono IR, Selli E, editors. Current trends in polymer photochemistry.

Hemel Hempstead: Ellis Horwood; 1995 [chapter 12].

[4] Mailhot B, Gardette J-L. Macromolecules 1992;25:4119.

[5] Mailhot B, Gardette J-L. Macromolecules 1992;25:4127.

[6] Grassie N, Weir NA. J Appl Polym Sci 1965;9:963.

[7] Rabek JF. Polymer photodegradation. London: Champman and Hall;

1995 [chapter 3].

[8] Rosık L. In: Svec P, Rosık L, Horak Z, Ve�cerka F, editors. Styrene-based

plastics and their modification. New York: Ellis Horwood; 1990 [chapter 6].

[9] David C, Piret W, Sekiguchi M, Geuskens G. Makromol Chem

1978;179:181.

[10] Hurum DC, Agrios AG, Gray KA, Rjah T, Thurnauer MC. J Phys Chem

B 2003;107:4545.

[11] Shang J, Chai M, Zhu Y. J Solid State Chem 2003;174:104.

Influence of transition metal-doped titanium(IV) dioxide on the photodegradation of polyethylene

Documents