than the larger ring 5g for the Phillips catalyst through a combination of
experiments and theoretical calculations [54, 114]. Recently, Zecchina et al.
reported the adsorption of probe molecules (CO, N 2 ) on the six-membered
chromasiloxane(II) ring 4g and found a good agreement between the calculated
vibrational frequencies and the experimental observations by increasing the percentage of HartreeÀFock exchange in the hybrid density functional B3LYP
[151]. Alternatively, we [71] studied the effects of Ti-modification of the Phillips
catalyst using the modified six-membered chromasiloxane ring 12g. Combined
with the experimental findings, a reasonable mechanistic understanding has been
made for the effects of Ti-modification of Phillips catalysts, such as promotion of
the polymerization activity, extension of MWD to the low molecular weight region,
and improvement of the distribution of inserted comonomers. Moreover, we [154]
studied the reduction of the hexavalent chromate species by ethylene during the
induction period and unraveled the behavior of Cr(II) active sites (4g) with or
without coordination of formaldehyde, which was generated through the reduction
of chromate species by ethylene. Recently, Tonosaki et al. [152] found that the
calculated activation energies for both ethylene insertion and chain transfer were in
good agreement using the six-membered chromasiloxane ring 4g and a slightly
larger cluster 13g. It was pointed out that the intrinsic origin of the broad MWD of
the polyethylene produced by Phillips catalysts might be derived from the multiple
coordination environments around the active Cr site on the silica surface.
It has long been recognized that the silica support is not an inert component of
the catalyst that simply directs polymer particle morphology. The neglect of the
real silica surface could introduce some artificial effects and provide an unrealistic environment for the adsorption of monomer on active chromium centers
[155]. Nowadays, with the improvement in computing resources and the development of quantum methodologies, full quantum calculations using a large
surface-supported model or a periodic model of silica gel surface can be
performed. Very recently, we developed a surface model 2d containing 37 Si
atoms through supporting of a six-membered chromasiloxane ring onto a silica
surface cutting from the β-cristobalite crystal structure, whose surface was
proposed to resemble that of amorphous silica [55]. The results were in good
agreement with the experimental spectra as discussed in Sect. 4. Guesmi and
Tielens [156] reported an amorphous silica surface slab containing 120 atoms
(Si 27 O 54 ·13H 2 O) that represented the amorphous character of the hydroxylated
silica surface involving different silanol types. Through a periodic DFT calculation, a higher stability of mono-oxo and di-oxo chromium species was confirmed in comparison with chromium-hydroxyl species. The main conclusion of
their study was a strong support of the six-membered chromasiloxane ring on the
amorphous silica surface as a valuable molecular model for the Phillips catalyst.
Thus far, all the related theoretical calculations mentioned above agree that the
six-membered chromasiloxane ring 4g could serve as a reasonable cluster model
for the Phillips catalyst, but that the effects of the silica support should be
considered as well.
180
R. Cheng et al.
experiments and theoretical calculations [54, 114]. Recently, Zecchina et al.
reported the adsorption of probe molecules (CO, N 2 ) on the six-membered
chromasiloxane(II) ring 4g and found a good agreement between the calculated
vibrational frequencies and the experimental observations by increasing the percentage of HartreeÀFock exchange in the hybrid density functional B3LYP
[151]. Alternatively, we [71] studied the effects of Ti-modification of the Phillips
catalyst using the modified six-membered chromasiloxane ring 12g. Combined
with the experimental findings, a reasonable mechanistic understanding has been
made for the effects of Ti-modification of Phillips catalysts, such as promotion of
the polymerization activity, extension of MWD to the low molecular weight region,
and improvement of the distribution of inserted comonomers. Moreover, we [154]
studied the reduction of the hexavalent chromate species by ethylene during the
induction period and unraveled the behavior of Cr(II) active sites (4g) with or
without coordination of formaldehyde, which was generated through the reduction
of chromate species by ethylene. Recently, Tonosaki et al. [152] found that the
calculated activation energies for both ethylene insertion and chain transfer were in
good agreement using the six-membered chromasiloxane ring 4g and a slightly
larger cluster 13g. It was pointed out that the intrinsic origin of the broad MWD of
the polyethylene produced by Phillips catalysts might be derived from the multiple
coordination environments around the active Cr site on the silica surface.
It has long been recognized that the silica support is not an inert component of
the catalyst that simply directs polymer particle morphology. The neglect of the
real silica surface could introduce some artificial effects and provide an unrealistic environment for the adsorption of monomer on active chromium centers
[155]. Nowadays, with the improvement in computing resources and the development of quantum methodologies, full quantum calculations using a large
surface-supported model or a periodic model of silica gel surface can be
performed. Very recently, we developed a surface model 2d containing 37 Si
atoms through supporting of a six-membered chromasiloxane ring onto a silica
surface cutting from the β-cristobalite crystal structure, whose surface was
proposed to resemble that of amorphous silica [55]. The results were in good
agreement with the experimental spectra as discussed in Sect. 4. Guesmi and
Tielens [156] reported an amorphous silica surface slab containing 120 atoms
(Si 27 O 54 ·13H 2 O) that represented the amorphous character of the hydroxylated
silica surface involving different silanol types. Through a periodic DFT calculation, a higher stability of mono-oxo and di-oxo chromium species was confirmed in comparison with chromium-hydroxyl species. The main conclusion of
their study was a strong support of the six-membered chromasiloxane ring on the
amorphous silica surface as a valuable molecular model for the Phillips catalyst.
Thus far, all the related theoretical calculations mentioned above agree that the
six-membered chromasiloxane ring 4g could serve as a reasonable cluster model
for the Phillips catalyst, but that the effects of the silica support should be
considered as well.
180
R. Cheng et al.