Unit 22  Unipol Process for Polyethylene
1. Process description
  In recent years, the UNIPOL process has become a popular commercial technology for linear polyethylene production (Burdett,1988). In this process, the copolymerization of ethylene and α-olefins is carried out in a fluidized-bed reactor using a heterogeneous Ziegler-Natta or supported metal oxide catalyst. A schematic diagram of the reactor system is shown in Fig.22.1. The feed to the reactor comprises ethylene, a comonomer (1-butene or a higher alpha-olefin),hydrogen and nitrogen. These gases provide the fluidization and heat transfer media and supply reactants for the growing polymer particles. The catalyst and a cocatalyst are fed continuously to the reactor. The fluidized particles disengage from the reactant gas in the expanded top section of the reactor. The unreacted gases are combined with fresh feed streams and recycled to the base of the reactor. Since the reaction is highly exothermic, heat must be removed from the recycle gas before it is returned to the reactor. The rate of polymer production is determined from an on-line heat balance. The mass of material in the bed is als
o calculated on-line using bed level and pressure measurements. The conversion per pass through the bed is very low, making the recycle stream much larger than the fresh feed streams. Because polymer particles in the fluidized bed are mixed well and the conversion per pass is low, gas composition and temperature are essentially uniform throughout the bed. Periodically, the product discharge valve near the base of the reactor opens and the fluidized product flows into a surge tank. The unreacted gas is recovered from the product that proceeds downstream for further processing and distribution.
The melt index and density of the polymer in the bed depend on catalyst properties, reactant gas
composition, and reactor temperature. The reactor is instrumented well with temperature, pressure, and flow sensors. Gas compositions are measured by on-line gas chromatographs. Melt index and density are measured every several hours in the quality control laboratory. These analyses require up to one hour. When the lab results become available, they are used to adjust the reactor operating conditions to ensure that on-specification polymer is produced.
2. Models for melt index and density
Any scheme to predict melt index and density between measurements requires a model describing how these variables are affected by reactor operating conditions. If the reactor is operated near one set of operating conditions to produce a limited number of polymer products, then an empirical linear plant model will often suffice. However, one of the advantages of the UNIPOL process over traditional liquid-phase systems is the wider range of products that can be produced (Burdett,1988). The models developed for this application must be valid over the range of products made in the reactor. Thus, linear empirical models are not suitable.
A Kinetic model describing molecular weight and copolymer composition development and their relationships to melt index and density is presented by Mc Auley et al (1990). While this model can predict MI, ρ, and production rate in an industrial reactor, the structure of this model (22 differential equations) is prohibitively complex for use in an on-line quality inference scheme. The approach taken in this article is to simplify the theoretical model so t
hat it becomes appropriate for on-line use. Although several different comonomers are used to produce linear polyethylene in UNIPOL systems, it is uncommon to operate with ethylene and more than two comonomers in the reactor simultaneously. Hence, the simple models for MI and ρ are developed for ethylene, butane and one higher alpha-olefin (HAO) comonomer. Extensions to more comonomers are straightforward.
Unmeasured impurities and unmodeled disturbances can result in sustained offset between model predictions and measured quality variables. If such drifts in product quality are not accounted for in the control scheme, then large quantities of off-grade polymer can be produced. One way to alleviate this problem is to force the model to track the process by updating parameters and predictions recursively on-line. If the common sources of the expected mismatch are known, then this information can be used to choose which parameters remain constant and which are likely to change due to the disturbances. Theoretically-based models have an advantage over empirical models in that the designer may have some prior knowledge about which parameters require on-line updating. Usually only a few meaningful parameters need to be updated, thereby making the on-line scheme
s easier to maintain and monitor.
翻译如下：
第22 单元 Unipol气相法聚乙烯生产工艺
1.工艺过程描述
近年来，对线性聚乙烯生产而言，UNIPOL工艺已经成为一个非常受欢迎的商业技术（伯德特，1988）。在此过程中，乙烯和α-烯烃在流化床反应器中进行共聚反应，使用了非均相的齐格勒—纳塔催化剂或负载化金属氧化物催化剂。图22.1所示的是一个反应器系统的示意图。反应器中的进料包括：乙烯，（1—丁烯或更高的长链α-烯烃）共聚单体，氢气和氮气。这些气体提供了流化、传热介质和供给反应物生长所需的聚合物颗粒。催化剂和助催化剂则被连续供给到反应器中。流化的颗粒在反应器中扩大的顶部部分脱离反应气体。未反应的气体与新鲜进料流相结合，并再循环到反应器的底部（基极）。由于该反应是高度放热的，热量必须在返回到反应器之前从循环气体中除去。聚合物的生产速率从在线的热平衡确定。通过床层高度和压力测量也可以在线计算反应器内的持固量。通过流化床的单程转化率
非常低，使得循环流远远大于新鲜进料流。因为在流化床中的聚合物颗粒是被充分混合的，而且单程转化率较低，所以整个流化床的气体组成和温度基本上是一致的。靠近反应器底部的出料阀打开，流化床产品流入缓冲罐，这一过程周期性地进行着。未反应的气体从产品中回收，产品进入下一工段进行进一步处理。
流化床中聚合物的熔融指数和密度取决于催化剂的性能，反应气体的组成，以及反应器的温度。该反应器是由温度、压力和流量传感器良好装备的，通过在线气相谱测定气体的组成，熔融指数和密度在质量控制实验室每隔几个小时测量一次。这些分析需要长达一小时。当实验的结果可行时，它们将被用来调整反应器的操作条件，以确保特定规格聚合物的生产。2.熔融指数和密度的测量模型
预测熔融指数和密度测量之间的任何计划，都需要一个模型来描述这些变量是如何受反应器操作条件影响的。如果反应器在相近的一组操作条件下被用来产生有限数目的聚合物产品，那么一个经验线性装置模型往往就足够了。然而，UNIPOL工艺优于传统液相系统的特点之一是能生产更宽范围的聚合物产品（伯德特，1988）。为此应用程序而开发的模型必须是在反应器中所制造产品的范围内有效，因此，线性经验模型是不适合的。

版权声明：本站内容均来自互联网，仅供演示用，请勿用于商业和其他非法用途。如果侵犯了您的权益请与我们联系QQ:729038198，我们将在24小时内删除。