vertical sheet of light (2mmthick). Two series of experimentswere
performed: (i) velocity measurements using PIV and (ii) concen-
tration measurements of the determining tracer using (PLIF). In
each series three different rotational impeller speeds: 225, 300
and 400 rpm,were employed. These impeller rotational speeds cor-
respond to impeller Reynolds numbers (Re = ND2/) of 37,500,
50,000 and 66,000, respectively. To determine of the average
velocity and mixing characteristics the CCD camera captured 500
instantaneous images in each case. The image capturing rate of the
CCD camerawas three frames per second. The average of these 500
images was used in the calibration of the PLIF measurement. The
calibration procedure is described by Moghaddas et al. (2002).
In the case of velocitymeasurements, the water in the tank was
seeded with Rhodamine-B fluorescent tracer particles, 1–20min
diameter. The flowfieldsweremeasured at 23◦,30◦ and 37.5◦ angles
behind the impeller blades for the three chosen rotational impeller
speeds, i.e. 225, 300 and 400 rpm, respectively. A CCD camera with
two long-pass filters, OG-550 and OG-570 (Melles Griot, Irvine, CA,
USA), were used to capture the fluorescent signals from the par-
ticles. The PIV image processing was performed using multipass
interrogation windows decreasing from 32×32 to 16×16 pixels.
The concentration andmixing timemeasurementswere carried
out using the PLIF technique.
In this case Rhodamine-590 was selected as the determining
tracer, as it has a highly emitted fluorescent intensity when irra-
diated with a 563 nm light and a linear relation to concentration
at low concentrations. The CCD camera (La Vision, FlowMaster 3S)
records fluorescence fromthe Rhodamine-590 tracer using two fil-
ters: an OG-570 with a cut-off of 570 nm (centre wavelength), and
an interference filter centred at 573 nm with 
 = 5 nm (Melles
Griot, Irvine, CA, USA).
The Rhodamine-590 solution was injected through an injection
port located belowthe liquid surface on a baffle parallel to the sheet
of laser light. The injection system consisted of a medical piston
pump that was easily controlled with regard to both the injection
speed and the flow rate. A Rhodamine-590 tracer solution of 0.2 g/l
with a constant flow rate 1.67ml/s was injected in 1.5 s into the
stirred tank. The change in the fluorescence signal from the tracer
with time, at chosen pixel positions in the stirred tank was then
measured. Prior to each mixing time measurement, the fluores-
cence signal of the bulk solution, without the determining tracer,
was measured to provide a background intensity. As the volume of
each injected pulse is less than 1/15,200 of the fluid in the tank,
it is assumed that the effect of the tracer on the flow pattern is
negligible.
The mixing time, t95, is defined as the time from ‘the release
of the determining tracer until the concentration of the tracer at a
particular pixel reaches 95% of the final concentration in the tank.
In practice, t95 is taken as the time, t, for the uniformity ratio of the摘要：现在计算和实验的方法已被广泛用于研究流场，我们在实验中用二个751叶涡轮搅拌桨在搅拌容器中充分混合运作。搅拌反应釜内有围绕旋转的叶轮叶片和静止的挡板之间液体流动的相互作用。在计算流体动力学（CFD）方法中，流场的计算用滑动网格（SM）的方法。紊流大涡模拟（LES）方法是用来模拟湍流。我们对仿真结果进行两个系列的实验验证：（1）利用粒子图像测速技术（PIV）对液相速度的测量及（2）使用平面激光诱导荧光（PLIF）技术用示踪剂确定被测物的液相浓度。每个系列中的叶轮有三种不同的转速：225rpm，300rpm和400 rpm。搅拌功率输入也是基于PIV技术的结果进行计算。运用这些技术，搅拌时间大幅度减少；增加叶轮转速的方法也增大了搅拌功率输入。这个令人满意的实验表明，CFD方法作为计算工具所拥有的潜力可以辅助设计搅拌反应器。 双Rushton搅拌反应釜的研究英文文献和翻译(3):http://www.751com.cn/fanyi/lunwen_8340.html