Propagation rate coefficients, k p , for free-radical polymerization of butyl acrylate (BA) previously reported by several groups are critically evaluated. All data were determined by the combination of pulsed-laser polymerization (PLP) and subsequent polymer analysis by size exclusion (SEC) chromatography. The PLP-SEC technique has been recommended as the method of choice for the determination of k p by the IUPAC Working Party on Modeling of Polymerization Kinetics and Processes. Application of the technique to acrylates has proven to be very difficult and, along with other experimental evidence, has led to the conclusion that acrylate chain-growth kinetics are complicated by intramolecular transfer (backbiting) events to form a mid-chain radical structure of lower reactivity. These mechanisms have a significant effect on acrylate polymerization rate even at low temperatures, and have limited the PLP-SEC determination of k p of chain-end radicals to low temperatures (<20 8C) using high pulse repetition rates. Nonetheless, the values for BA from six different laboratories, determined at ambient pressure in the temperature range of À65 to 20 8C mostly for bulk monomer with few data in solution, fulfill consistency criteria and show excellent agreement, and are therefore combined together into a benchmark data set. The data are fitted well by an Arrhenius relation resulting in a preexponential factor of 2.21 Â 10 7 L Á mol À1 Á s À1 and an activation energy of 17.9 kJ Á mol À1 . It must be emphasized that these PLP-determined k p values are for monomer addition to a chain-end radical and that, even at low temperatures, it is necessary to consider the presence of two radical structures that have very different reactivity. Studies for other alkyl acrylates do not provide sufficient results to construct benchmark data sets, but indicate that the family behavior previously documented for alkyl methacrylates also holds true within the alkyl acrylate family of monomers.Arrhenius plot of propagation rate coefficients, k p , for BA as measured by PLP-SEC.

A novel method to extract individual free-radical polymerization rate coefficients for butyl acrylate intramolecular chain transfer (backbiting), k bb , and for monomer addition to the resulting midchain radical, k p t , is presented. The approach for measuring k bb does not require knowledge of any other rate coefficient. Only the dispersion parameter of SEC broadening has to be determined by fitting measured MWDs or should be available from separate experiments. The method is based upon analysis of the shift in the position of the inflection point of polymer molecular weight distributions produced by a series of pulsed-laser polymerization (PLP) experiments with varying laser pulse repetition rate. The coefficient k bb is determined from the onset of the sharp decrease of the apparent propagation rate coefficient (k p app ) with decreasing repetition rate, an approach verified by simulation. With experiments performed between -10 and +30°C, the estimated values are fitted well by an Arrhenius relation with pre-exponential factor A(k bb ) ) (4.84 ( 0.29) × 10 7 s -1 and activation energy E a (k bb ) ) (31.7 ( 2.5) kJ‚mol -1 . At low pulse repetition rates, the experimental k p app values are related to an averaged propagation rate coefficient, k p av , that is dependent on the relative population of chain-end and midchain radicals. Evaluated by comparing simulated and experimental molecular weight distributions, k p av provides an estimate for k p t . The Arrhenius parameters are: A(k p t ) ) (1.52 ( 0.14) × 10 6 L‚mol -1 ‚s -1 and E a (k p t ) ) (28.9 ( 3.2) kJ‚mol -1 .

New expressions that account for the formation of acrylate midchain radicals by intramolecular transfer and their subsequent propagation, termination, and transfer events have been derived for polymerization rate, average chain-length, and chain-length distribution under stationary conditions. The nonidealities observed in previous kinetic studies are captured in a single lumped rate coefficient, θ, that controls the apparent order of rate on monomer concentration. Applied to rate data from the literature, the treatment yields consistent estimates for θ and k p/kt 0.5 for butyl acrylate polymerization at 50°C. Combining these ratios with chain-end propagation values determined by pulsed-laser polymerization, intramolecular transfer rate coefficients estimated from 13 C NMR data, and/or radical concentrations measured by ESR provides a new means to estimate the individual rate coefficients for acrylate polymerization systems. It is also shown that estimates for butyl acrylate transfer to monomer rate coefficients obtained from chain-length distributions are valid even in the presence of acrylate backbiting events.

Unusual difficulties are faced in the determination of propagation rate coefficients (kp) of alkyl acrylates by pulsed‐laser polymerization (PLP). When the backbiting is the predominant chain transfer event, the apparent kp of acrylates determined in PLP experiments for different frequencies should range between kp (propagation rate coefficient of the secondary radicals) at high frequency and k pav at low frequency. The k pav value could be expressed from kinetic parameters: $k_{\rm p}^{{\rm av}} = k_{\rm p} - {{k_{\rm p} - k_{{\rm p}2} } \over {1 + {{k_{{\rm p}2} [{\rm M}]} \over {k_{{\rm fp}} }}}}$, where kfp is the backbiting rate coefficient, kp2 is the propagation rate coefficient of mid‐chain radicals, and [M] is the monomer concentration.

n-Butyl acrylate (BA) starved-feed solution semibatch experiments with varying final polymer content and monomer feed times were carried out at 138 °C. A full mechanistic model of the system implemented in Predici includes intermolecular chain transfer to polymer and macromonomer propagation as well as backbiting, chain scission, and midchain radical propagation and termination. The importance of macromonomer propagation under these conditions of industrial interest is illustrated by experiment and simulation, with the macromonomer reaction responsible for the significant increase in polymer weight-average molecular weight ($\overline M _{\rm w}$) with time. Rate coefficients for macromonomer propagation (k(mac) ) and β-scission (k(β) ) of k(mac) /k(p)  = 0.55 and k(β)  = 12 s(-1) (with k(p) the rate coefficient for BA chain-end propagation) provide a good representation of experimental $\overline M _{\rm w}$ and macromonomer end group data at 138 °C.

The kinetics of butyl acrylate solution polymerization up to high conversion are studied both experimentally and theoretically. New expressions taking into account both inter‐ and intra‐molecular transfer to polymer are derived for polymerization rate, average chain‐length and branching level. These expressions are used for comparative analysis of these two chain transfer events and for evaluation of the backbiting rate coefficient, kbb, and the constant of chain transfer to solvent (xylene), CtrS. Using batch experiments performed at 50, 60 and 70 °C, the Arrhenius parameters for kbb and CtrS are estimated to be, respectively, A(kbb) = (7.4 ± 1.5) × 107 s−1, Ea(kbb) = (32.7 ± 0.5) kJ · mol−1 and A(CtrS) = 10.5 ± 3.5, Ea(CtrS) = (26.9 ± 1.9) kJ · mol−1.

Butyl acrylate solution polymerization has been studied at 80-170 8C over a range of monomer and initiator levels. A PREDICI model has been developed to describe the experiments, taking into account chain backbiting, b-scission, and addition of macromonomer. The model is used to estimate Arrhenius parameters for the rate coefficients of b-scission [A(k b ) ¼ (1.49 AE 0.28) Â 10 9 s À1 , E a (k b ) ¼ 63.9 AE 0.9 kJ Á mol À1 ], for the ratio of the termination rate coefficients of secondary and midchain radicals Aðk t =k t t Þ ¼ 0:74 AE 0:08; E a ðk t =k t t Þ ¼ Àð11:2 AE 0:5ÞkJ Á mol À1 Â Ã and for the constant of chain transfer to xylene [A(C trS ) ¼ 107 AE 13, E a (C trS ) ¼ 35.4 AE 0.8 kJ Á mol À1 ] by fitting simulated dependencies to the experimental data.

The Outokumpu Scientific Deep Drill Hole intersects a 2500 m deep Precambrian crustal section comprising a 1300 m thick biotite-gneiss series (mica schists) at top, followed by a 200 m thick meta-ophiolite sequence, underlain again by biotite gneisses (mica schists) (500 m thick) with intercalations of amphibolite and meta-pegmatoids (pegmatitic granite). From 2000 m downward the dominating rock types are metapegmatoids (pegmatitic granite). Average isotropic intrinsic P-and S-wave velocities and densities of rocks were calculated on the basis of the volume fraction of the constituent minerals and their single crystal properties for 29 core samples covering the depth range 198 m 2491 m. The modal composition of the rocks is obtained from bulk rock (XRF) and mineral chemistry (microprobe), using least squares fitting. Laboratory seismic measurements on 13 selected samples representing the main lithologies revealed strong anisotropy of P-and S-wave velocities and shear wave splitting. Seismic anisotropy is strongly related to foliation and is, in particular, an important property of the biotite gneisses, which dominate the Upper and Lower gneiss series. At in situ conditions, velocity anisotropy is largely caused by oriented microcracks, which are not completely closed at the pressures corresponding to the relatively shallow depth drilled by the borehole, in addition to crystallographic preferred orientation (CPO) of the phyllosilicates. The contribution of CPO to bulk anisotropy is confirmed by 3D velocity calculations based on neutron diffraction texture measurements. For vertical incidence of the wave train, the in situ velocities derived form the lab measurements are significantly lower than the measured and calculated intrinsic velocities. The experimental results give evidence that the strong reflective nature of the ophiolite-derived rock assemblages is largely affected by oriented microcracks and preferred crystallographic orientation of major minerals, in addition to the lithologic control.