5 Reactor Design-reactor type and configuration
化工术语中英文
ammonia in excess ammonia liquefier ammonia plant ammonia product pump ammonia pump ammonia synthesis ammonium bicarbonate water stripper ammonium carbomate ammonium carbomate dehydrating to form urea ABC：ammonium carbonate AN：acryionitrile Analog instrument Analysis manual Analyzer cabinet Analyzer house Analyzer shelter Analyzers - Oxygen Zirconium Oxide Cell anchor bolt angel hair angle stair angle valve angles anion exchanger anit-sweat insulation annular chamber annular flow another train in spare antioxidant antiseptics antisurge, anti-surge anti-sweat antiswirl baffle (anti-swirl baffle) anti-whip anti-whip device anti-whip restraints AO：Analog Output APC：Advanced Process Control API：American Petroleum Institute Annex aqueous emmissions Ar：Argon arenes aromatics As：arsenic (As) arsine removal bed asbestos ash melting point ash treatment aspect ratio aspirator air assembly drawing Asset Management System(AMS) ASU：Air Seperation t
沈大中-变压器有载分接开关原理和应用1
7. 分接操作程序
MR R&D SHen
Date
Page 14
8.1
操作程序– 电阻式切换开关 (OILTAP M)
I
Ust R R R R R R R R
I
R
R
R
R
R
R
R
R
main contact main switching contact transition contact R transition resistor
• Diverter switch + tap selector (AVT, VT)
Date Page 8
MR R&D SHen
6.1
油浸式有载分接开关 (OILTAP)
OILTAP R I 2002
Connection for the protective device RS Insulation to ground Diverter switch, transition resistor Tap selector
P1, P4: tap selector contacts P2, P3: by-pass switch contacts VI vacuum interrupter
Date Page 16
9. 电抗式和电阻式有载分接开关的比较
Resistor type Mechanic stress: Inspection interval: Highest voltage for equipment Um: Maximum voltage tap to tap: Maximum through-current: No. of tap selector contacts: No. of operating positions: Motor drive running time per operation: Temperature range of oil: - service in arctic regions high 40,000 .. 150,000 op. 72.5 .. 362 kV 1000 .. 4000 V 300 .. 3000 A 9, 10, 12, 14, 16, 18, 22 up to ± 17 2) app. 5.4 s -25 °C .. +105 °C - 40 °C (V, M, R) Reactor type 1) low 500,000 op. 15 .. 69 kV 2000 V 600 .. 2500 A 9 ± 16 ( x 1000 V 3)) app. 2 s 4) - 25 °C .. + 105 °C - 40°C 5)
高温气冷反应堆设计的新概念
髙温气冷反应堆设计的新概念田嘉夫（清华大学核鞫支术设计研究院，北京100084 ）摘要：高温气冷堆能够髙效率发电和供应髙温工艺用热。
英模块化设汁有两种类型，即柱状燃料堆和球形燃料堆。
本文提岀了一种新颖的模块式高温气冷堆的概念设讣，它的堆芯是由燃料球的规则堆积床构成，不同于现有的这两种堆型。
规则床中的燃料球在平而上成正方形排列，四个球的中心是次一层球的位置，形成正四棱锥堆积。
肖燃料球从顶部落入被做成一立几何形状的堆芯空腔时，就形成规则堆积。
同样燃料球也将从顶部开孔卸出。
规则床堆与其它模块化设计一样，其堆芯形状能够允许向外界非能动冷却，并且保持最高燃料温度在安全范愠内。
但规则床堆却能够提髙输出功率和降低堆芯压降，同时兼有球形燃料堆和柱状燃料堆的主要优点。
关键词：模块式髙温气冷堆；卵石床；规则床；球形燃料堆；柱状燃料堆A Novel Concept of the High Temperature Gas Cooled Reactor DesignTian Jiafu(Institute of nuclear energy technology, Tsinghua University, Beijing 100084,China)Abstract: The High Temperature Gas Reactor (HTGR) can offer a high-efficiency electricity production and a broad range of process heat application. The reactor core type of the modular design can be a prismatic block or a pebble bed core・ This paper presents a novel conceptual design of the modular HTGR in which the reactor core is filled with an ordered packing bed of fuel spheres and different from the existing two types. The ordered beds are packed in a pyramid geometry in which the unit cell layer is formed by four fuel spheres lying at the corners of a square. and the individual spheres in subsequent layers fill the cusps formed by them・ This arrangement allows fuel elements to be poured into the core cavity which is shaped so that an ordered bed is formed and to be discharged from the core through the opening holes in the reactor top. The core geometry of the ordered bed reactor as a modular design is such that passive cooling to the environment and maxinnim fuel temperatures are kept within safe limits・ However the ordered bed reactor is allowed to increase reactor output power and decrease core pressure drop as well as having most of the advantages of both the pebble bed reactor and block type reactor.Key words：Modular HTGR: pebble bed; ordered bed; pebble bed reactor; block type reactor1.高温气冷反应堆的技术进展气体冷却反应堆与水冷却反应堆一样是最早开发研究的堆型之一。
催化裂化
13
Fluidised Catalytic Cracking
FCC Catalyst. Zeolite
The main active component in the catalyst is the Y-Zeolite. Y-Zeolite is a crystalline structure of aluminosilicates which has the
Figure 3. Structure of Y-faujasite
14
Fluidised Catalytic Cracking
FCC Catalyst. Zeolite
In the cracking of long chain paraffins, a type of high silica zeolite ZSM-5 is added.
paraffins olefins naphthenes
side chains in aromatics
Название темы
Figure 2. FCC reactions network
6
Fluidised Catalytic Cracking
Primary Reactions
Primary cracking occurs by the carbenium ion intermediates in the
The high endothermic nature of cracking reactions requires that the reactor operates at high temperatures 480–550 ºC.
Scylla数据库性能最大化最佳实践指南说明书
BEST PRACTICES GUIDEGet me up and runningIn case you are not able to read this document in full, here are the most important things to remember:Why should I read this? I already know how to execute a benchmarkScylla is different from any other NoSQL database. It achieves the highest levels of performance and takes full control of the hardware by utilizing all of the server cores in order to provide strict SLAs for low-latency operations. If you run Scylla in an over-committed environment, performance won’t just be linearly slower — it will tank completely.This is because Scylla has a reactor design that runs on all the (configured) cores and a scheduler that assumes a 0.5 ms tick. Scylla does everything it can to control queues in userspace and not in the OS/drives. Thus it assumes the bandwidth that was measured by scylla_setup.However, it is not difficult to get the best performance out of Scylla. It primarily tunes itself automatically. Just make sure you don’t work against the system.Install Scylla Monitoring StackInstall and use the Scylla Monitoring Stack, which provides excellent additional value above and beyond performance optimization. If you cannot pinpoint a performance bottleneck, you likely have not configured the system correctly. Scylla Monitoring Stack will help to sort this out.With the recent addition of the Scylla Advisor to the Scylla Monitoring Stack, it is now even easier to find potential issues.Install Scylla ManagerInstall and use Scylla Manager together with the Scylla Monitoring Stack. Scylla Manager provides automated backups, and repairs of your database. Scylla Manager can manage multiple Scylla clusters and run cluster-wide tasks in a controlled and predictable way.Run scylla_setupBefore running Scylla, it is critical that the scylla_setup script has been executed. Scylla doesn’t require manual optimization – it is the task of the scylla_setup script to determine the optimal configuration. If scylla_setup has not run, the system won’t be configured optimally.Read more here.Coordinated omission Read more here.Read more here.possible. Reproducible resultsQuery recommendationsCorrect data modelingThe key to a well performing system is using the properly defined data model. A poorly structured data model can easily lead to an order-of-magnitude performance difference compared to a proper model.A few of the most important tips:• Choose the right partition key and clustering keys. Reduce or even eliminatethe amount of data that needs to be scanned.• Add indexes where appropriate.• Partitions that are accessed more than others (hot partitions) should beavoided because they cause load imbalances between CPUs and nodes.• Large partitions, large rows and large cells should be avoided because theycan cause high latencies.Use prepared statementsPrepared statements provide better performance because: parsing is doneonce, token/shard aware routing and less data is sent. Apart from performance improvements, prepared statements also increase security because they prevent CQL injection.Read more here.Use paged queriesIt is best to run queries that return a small number of rows. But if a query could return many rows, then an unpaged query can lead to a huge memory bubble and Scylla could eventually decide to kill the query. With a paged query, the execution collects a page’s worth of data and new pages are retrieved on demand, leading to smaller memory bubbles.Read more here.Don’t use reverse queriesWhen using a query with an ORDER BY clause, you need to make sure that the order is the same as in the data model. Otherwise you run into a problem called reverse queries, which can cause unbound memory usage and killed queries.Use workload prioritizationIn a typical application there are operational workloads that require low latency. Sometimes these run in parallel with analytic workloads that process high volumesof data and do not require low latency. With workload prioritization, one can prevent the analytic workloads from negatively impacting the latency-sensitive operational workload.Read more here.Bypass cacheThere are certain workloads, e.g. analytical workloads, that scan through all the data. By default Scylla will try to use cache, but since the data won’t be used again, it leads to cache pollution — good data is pushed out of the cache and replaced by useless data.This can result in bad latency on operational workloads due to increased rate of cache misses. T o prevent this problem, queries from analytical workloads can bypass the cache using the ‘bypass cache’ option.Read more here.BatchingMultiple CQL queries to the same partition can be batched into a single call. Imagine the round trip time being 0.9 ms and the service time time 0.1 ms. Without batching the total latency would be 10x(0.9+0.1)=10.0 ms. But if you would create a batch of 10 instructions, the total time would be 0.9+10*0.1=1.9 ms. That is 19% of the latency compared to no batching.Read more here.Driver guidelinesUse the Scylla drivers that are available for Java/Python/Go. They provide much better performance than third-party drivers because they are shard aware – theycan route requests to the right CPU core (shard). When the driver starts, it gets the topology of the cluster and therefore it knows exactly which CPU core should get a request.If Scylla drivers are not an option, make sure that at least a token-aware driver is used so one round trip is removed.Check if there are sufficient connections created by the client, otherwise performance could suffer. The general rule is between 1-3 connections per Scylla CPU per node.Read more here.handling, etc.Cloud compute instance recommendationsScylla is designed to utilize all hardware resources. Bare metal instances are preferred for best performance.Read more here.Image guidelinesUse Scylla provided AMI on AWS EC2, if possible. They have already been correctly configured.AWSAWS EC2 i3, i3en and cd5 bare metal instances are highly recommended because they are optimized for high I/O.Read more here.If bare metal isn’t possible, use Nitro-based instances and run with ‘host’ as tenancy policy. This will prevent the instance being shared with other VMs.If the recommendation above isn’t possible, we recommend instance storage over EBS. If instance store is not an option, use an io2 IOPS provisioned SSD for best performance. If there is limited support for instance storage, place the commitlog there. There is a new instance type available called r5b that has high EBS performance.Read more here.GCPFor GCP we recommend n1/n2-highmem with local SSDs.Read more here.AzureFor Azure we recommend the Lsv2 series. They feature high throughput and low latency and have local NVMe storage.Read more here.DockerWhen running in the Docker platform, please use CPU pinning and host networking for the best performance.Read more here.KubernetesAs with Docker, CPU pinning should be used on Kubernetes environments as well. In this case the pod should be pinned to a CPU so that no sharing takes place.Consistency LevelThe consistency level determines how many nodes the coordinator should wait for in order for the read or write to be considered a success. The consistency level is determined by the application based on requirements for consistency, availability and performance. The higher the consistency, the lower the availability and the performance.For single data center setups a frequently used consistency level for both reads and writes is QUORUM. It gives a nice balance between consistency and availability/ performance. For multi-datacenter setups it is best to use LOCAL_QUORUM.Read more here.Replication FactorThe recommended replication factor is set to 3, and in most cases this is a sensible default because it provides a good balance between performance and availability. Keep in mind that a write will always be sent to all replicas, no matter the consistency level.Scylla has excellent performance out of the box. Following the best practices described in this paper will prevent mistakes that might diminish the performance of your Scylla deployment.United States Headquarters2445 Faber Place, Suite 200 Palo Alto, CA 94303 U.S.A. Israel Headquarters 11 Galgalei Haplada Herzelia, IsraelABOUT SCYLLADBScylla is the real-time big data database. API-compatible with Apache Cassandra and Amazon DynamoDB, Scylla embraces a shared-nothing approach that increases throughput and storage capacity as much as 10X. Comcast, Discord, Disney+ Hotstar, Grab, Medium,Starbucks, Ola Cabs, Samsung, IBM, and many more leading companies have adopted Scylla to realize order-of-magnitude performance improvements and reduce hardware costs. Scylla’s database is available as an open source project, an enterprise edition and a fully managed database as a service. ScyllaDB wasfounded by the team responsible for the KVM hypervisor. For more information: 。
第5章 间歇式反应器05
• 3. 衡算方程 必须在控制体积内进行。 • 物料衡算： • 积累＝进入－流出＋产生－消耗 • 能量衡算： • 积累＝输入－输出＋产生－消耗
5.2 间歇操作搅拌槽式反应器 （BSTR）
• Batch Stired Tank Reactor • 间歇操作搅拌槽式反应器的操作时间系由反应时间和辅助时间 两部分组成。 • 反应时间 tr，即开始进行反应直到达到所要求的反应程度为止 所需要的时间，常以 tr表示，它的大小与该反应的动力学与所 要求的反应程度有关；可通过动力学模型进行计算。， • 辅助操作时间tb，包括装料、灭菌、卸料、清洗等所需时间之 总和，以tb表示。tb是根据生产经验来确定的。 • 间歇搅拌槽式反应器，有两个主要特性：一是在反应进行过程 中无物料的输人和输出；二是反应器内物料充分混合，浓度、 温度均一，而且反应物系的浓度仅随反应时间而变化。因此可 以对整个反应器做物料平衡。
tr
（5－17）
LVR
tr
dCS (1 L )VRrS dt
L C 1 K m ln S 0 CS 1 L rmax
（5－21）
• 对于内扩散阻力可忽略的M－M方程：
• 积分:
L 1 L
X
1 dC CS0 r S
C
S
S
1 C tr L [(CS0 CS ) Km ln S0 1L rmax CS
• 3）大型化生物反应器的开发研究。 生物反应 器正向大型化方向发展。例如：生产抗生素的 发酵罐容积已达400m3，氨基酸的达300m3， 生产单细胞蛋白的气升式发酵罐达2300m3， 处理废水的生化反应器的容积甚至超过 27000m3，国内生物反应器的容积多在200m3 以下。反应器的放大降低了生产成本，但大型 反应器的设计还存在一定的技术问题亟待解决； • 4）特殊要求的新型生物反应器的研制开发。 ）特殊要求的新型生物反应器的研制开发 如基因产品生产、细胞固定化及动植物细胞培 养的工业反应器，固体发酵反应器、边发酵边 分离反应器等的开发研制。 • 5）反应条件的检测与自动控制。
朴克自动炉工程公司产品介绍说明书
MAWP: 2,200 psi @ 450 °F (151 bar @ 232 °C)
Volumes: 500 ml 1,000 ml 2,000 ml 4,000 ml
Seal Materials (Temp): Nitrile (250 ºF / 121 ºC) Ethylene-Propylene (300 ºF / 149 ºC) PTFE (400 ºF / 204 ºC) Fluorocarbon (450ºF / 232ºC) Silicone (400 ºF / 204 ºC) Perfluoroelastomer (500 ºF / 260 ºC)
Vessel Materials: ANSI 316 Stainless Steel Hastelloy® C-276
Notes: MAWP based upon Fluorocarbon seals, temperature and pressure ratings will vary depending upon seal material selected. See Bulletin “PV-ZIP” for details of vessel performance. ASME Code Stamp not available for the Zipper Closure. Alternative materials are available, please consult factory.
ZipperClave®
EZE-Seal
Bolted Closure
The ZipperClave® offers the quickest opening autoclave available. No bolts to torque, nor clamps or rings. Inserting a single spring section closes the vessel. The ZipperClave® uses an elastomeric O-ring for it’s pressure seal.
专业英语翻译(9.13.14.15.16.23)
Unit 91、As is found in all phases of polymer chemistry, there aremany exceptions to this categorization.译：正如在高分子化学的各个部分都可以看到的那样，对这种分类方法有很多例外情况。
2、When the molecular chains are ‘straightened out’orstretched by a process of extension, they do not have sufficient attraction for each other to maintain the oriented state and will retract once the force is released. 译：当通过一个拉伸过程将分子链拉直的时候，分子链彼此之间没有足够的相互吸引力来保持其定向状态，作用力一旦解除，将发生收缩。
3、Therefore, a potential fiber polymer will not become a fiberunless subjected to a ‘drawing’process, i.e., a process resulting in a high degree of intermolecular orientation. 译：因此，可以制成纤维的聚合物将不成其纤维，除非经受一个抽丝拉伸的过程，即一个可以形成分子间高度取向的过程。
4、It must be borne in mind that, with the advent ofZiegler-Natta mechanisms and new techniques to improve and extend crystallinity, and the closeness of packing of chains, many older data given should be critically considered in relation to the stereoregular andcrystalline structure.译：必须牢牢记住，随着Ziegler-Natta机理出现，以及随着提高结晶度和提高链的堆砌密度的新方法的出现，对许多过去已得到的关于空间结构和晶体结构旧的资料，应当批判的接受。
Design and additive manufacture for flow chemistry
Cite this:Lab Chip ,2013,13,4583Received 16th July 2013,Accepted 31st August 2013DOI:10.1039/c3lc50844g /locDesign and additive manufacture for flow chemistry †Andrew J.Capel,a Steve Edmondson,*a Steven D.R.Christie,*b Ruth D.Goodridge,c Richard J.Bibb d and Matthew Thurstans dWe review the use of additive manufacturing (AM)as a novel manufacturing technique for the production of milli-scale reactor systems.Five well-developed additive manufacturing techniques:stereolithography (SL),multi-jet modelling (MJM),selective laser melting (SLM),laser sintering (LS)and fused deposition modelling (FDM)were used to manufacture a number of miniaturised reactors which were tested using a range of organic and inorganic reactions.IntroductionWith continuous flow chemistry gradually being integrated into laboratories,considerable research and development has been focused on innovative manufacturing methods for reac-tor systems.1A number of manufacturing technologies for both millilitre-scale and microlitre-scale flow reactors are currently commercially used.Microlitre-scale reactors typically take the form of chips of inert polymer (e.g.PDMS)or glass with a two-dimensional pattern of flow channels.2These are manufactured using a range of well-developed batch and mass-production techniques such as hot embossing,3laser ablation,4micro-machining 5or chemical etching,6and can feature intricate designs such as mixing pathways,online analysis and temperature-controlled zones.These techniques are generally constrained in design to two-dimensional planar channel networks,with more intricate designs leading to significant increases in cost,manufacturing complexity and production time.In contrast,millilitre-scale reactors are often simply lengths of polymer (e.g.PTFE)and metal (e.g.stainless steel)tubing,7or large glass chips etched with relatively sim-ple channels.Although tubing is commercially available and cheap,it does not allow for the design complexity which has proven possible in microfluidics,such as the incorporation of intricate mixing pathways.8In this paper,we explore the use of additive manufacturing (AM)as a method of fabricating millilitre-scale flow reactors,to allow the incorporation of the innovative designs which have previously been limited to microlitre-scale microfluidic devices.Developed from rapid prototyping,additive manufactur-ing,also known as additive layer manufacturing or ‘3D printing ’,is the umbrella term used to cover a variety of technologies that can directly produce complex three dimensional parts,with near-complete design freedom.9The term additive manufacturing is now widely accepted and is recognised by international standards.10In AM,parts are built layer-by-layer,using techniques such as extrusion,11jetting,12powder sintering 13or photo-polymerisation.14Parts can be made with complex and customizable geometries,including those diffi-cult to make by conventional manufacturing techniques.15Although still a rapidly developing area of research,AM already offers a range of manufacturing techniques capable of meeting many of the design criteria previously outlined.The design freedom offered by AM can allow the produc-tion of bespoke reactors with complete control over flow path length and volume,and therefore the reaction residence time for each reactor.More complex internal geometries such as turbulent and static mixers are also well within the scope of AM mercially,AM can also offer consider-able benefits in terms of build cost and production speed.AM manufactured parts can often be taken from concept to realisation within hours,with huge time-savings possible in both the design and manufacturing processes.Manufactur-ing costs for complex parts such as miniaturised reactor systems can be substantially reduced,with the cost of pro-duction directly related to the volume of material used,rather than the necessity for specialist tooling and labour.16This is exemplified by the recent work of Cronin,who has produced “reactionware ”using a basic extrusion-based AM system.17–20However,using AM in chemistry also poses some funda-mental challenges.Although the range of materials AM offers is growing,most AM processes are dedicated or limited to specific material types.The chemical inertness,thermal stability and transparency of glass make it desirable foraDepartment of Materials,Loughborough University,Loughborough,LE113TU,UK.E-mail:S.Edmondson@ bDepartment of Chemistry,Loughborough University,Loughborough,LE113TU,UK.E-mail:S.D.Christie@ cAdditive Manufacturing and 3D Printing Research Group,Nottingham University,Nottingham,NG72RD,UK.E-mail:Ruth.Goodridge@ dDesign School,Loughborough University,Loughborough,LE113TU,UK.E-mail:R.J.Bibb@†Electronic supplementary information (ESI)available:Further experimental details and images of reactors RD1–RD9are available in the supplementary information.See DOI:10.1039/c3lc50844gLab on a ChipP u b l i s h e d o n 03 S e p t e m b e r 2013. D o w n l o a d e d b y X i a m e n U n i v e r s i t y o n 17/11/2015 11:46:59.chemical reactor manufacture;however,the processing of glass by AM is still at the research stage and there is not currently a commercial process capable of producing glass components with the mechanical properties and build reso-lution required for this application.In addition to material limitations,each AM technique has differing limitations of allowable part geometries,accuracy,and spatial resolution.For a given reactor design,this will lead to trade-offs in the selection of AM process and material.In this paper,we demonstrate the applicability of five common AM techniques to millilitre-scale reactor fabrication:stereolithography,multi-jet modelling,selective laser melt-ing,selective laser sintering and fused deposition modelling.We have considered the strengths and weaknesses of each process,whilst highlighting the limited prior work in the application of AM to reactor design.We also consider the strengths and weaknesses of existing micro-scale and nano-scale reactor manufacturing technologies,and discuss how innovative additive manufacturing techniques could be com-plementary to the world of lab-on-a-chip.Results and discussionFused deposition modellingFused deposition modelling is one of a number of commer-cially available and relatively inexpensive extrusion-based AM manufacturing techniques.FDM uses a temperature-controlled extrusion nozzle to deposit a viscous molten ther-moplastic polymer bead into a three-dimensional structure in a layer-by-layer process.21FDM can use a number of suitably inert thermoplastics such as acrylonitrile butadiene styrene (ABS),polycarbonate (PC),polycarbonate-acrylonitrile butadi-ene styrene (PC-ABS)and polyetherimides such as ULTEM.Two recent studies have seen silicone and polypropylene reactors made using commercially available and relatively cheap,extrusion based AM systems.The Fab@Home 22and 3DTouch 23systems utilised in the studies,use a syringe-based extrusion nozzle to deposit the material into a pre-designed 3-D geometry.The first study used a Fab@Home extrusion system to manufacture a centimetre-scale acetoxysilicone poly-mer (LOCTITE bathroom sealant)reagent silo.The silo was used to store and deliver reagents into a sealed reaction chamber whilst maintaining the structural integrity of the reactor.The reactor was used to undergo a range of inor-ganic cluster formations along with an organic heterocyclic ring formation.18The second study used a 3DTouch extru-sion system to manufacture a range of milli-scale and micro-scale chip reactors.17The reactors were manufactured using polypropylene,a thermoplastic polymer inert to a range of mild organic solvents and reagents,and attached to PTFE tubing using a commercially available epoxy resin.The reac-tors were used to undertake a range of reactions in flow including an imine formation,an alkylation reaction and a range of inorganic self-assembly reactions.Very recently,Cronin and co-workers have coupled an AM reactor to a mass spectrometer for analysis (using conventional flow chemistrycoupling hardware)19and demonstrated the high degree of design freedom in AM by fabricating a compartmentalised reactor in which synthesis steps are initiated by simply rotating the device.20These recent studies successfully demonstrated the appli-cability of extrusion-based AM procedures towards the manu-facture of customisable milli-scale and macro-scale reactors.Whilst comparatively cheap,these extrusion processes are crude,typically intended for hobbyists or less demanding prototyping applications and are therefore limited in dimen-sional accuracy and build resolution when constructing com-plex miniaturised geometries.The current range of materials available is limited and those available are generally not suit-able for use across a wide range of organic reactions.We decided to investigate the potential of FDM as a manufacturing process by preparing a relatively complex part,a split-and-recombine (SAR)static mixer,from ABS using a commercial Stratasys Dimension 3D Modelling Printer.24The virgin (non-recycled)feedstock materials used were as supplied by the manufacturer in sealed cartridges,and have a long shelf-life due to the environmental stability of these widely-used thermoplastics.This type of mixing technology is well developed in micro-scale chemistry,and offers sub-stantial benefits in terms of reaction rate and yield.25This advanced FDM machine enables the use of a secondary solu-ble supporting material that can be completely removed from the main part after building by immersion in an agitated detergent plex internal channels can be built that would not be possible with the simpler,cheaper 3D printers used in the previous studies described above.The reactor design (RD1)consisted of a tube of length 510mm,diameter 3mm and a total volume of 3.6mL,contained within a wall thickness of 5mm.Post-manufacture machining of a standard screw thread on the inlet and outlet allowed connection to a commercially available FlowSyn continuous flow system.Water was flowed through the reactor at 1mL min −1and 20bar pressure;however numerous leaks occurred along the flow path.It was surmised that the extrusion based manufacturing process had not produced a fully dense tube wall capable of holding water under high pressure.It is highly probable that the integrity of the wall fails along the inter-layer boundary (Fig.1).The accuracy and resolution of FDM is limited to the bead size deposited,which also dictates the layer thickness.Inter-layer bonding depends on localised re-melting of the previous layer by the deposited layer and consequently it is typical for FDM parts to have poorer physical properties perpendiculartoFig.1FDM split-and-recombine mixer (RD1).Lab on a ChipPaperP u b l i s h e d o n 03 S e p t e m b e r 2013. D o w n l o a d e d b y X i a m e n U n i v e r s i t y o n 17/11/2015 11:46:59.the build plane.26On small parts accuracies of around ±0.3mm are achievable in practice.Typically dimensional accuracy in the XY build plane can be good approaching ±0.1mm using the most sophisticated machines.However,accuracy in the Z direction is a function of layer yer thickness can vary from around 0.35to 0.1mm depending on the exact machine specification and conse-quently achievable accuracy will be more like ±0.25mm even when using sophisticated machines.In this example the layer thickness was 0.25mm,which is comparatively large com-pared to other professional standard AM processes.FDM machines are mechanically robust and repeatability is gener-ally good.It may be possible to improve part density and build resolution by using a machine with a smaller layer thickness such as the Stratasys 400mc,as well as altering build parame-ters such as raster patterns and nozzle diameter.However,AM techniques that demonstrate strong inter-layer bonding and higher build resolution than the bead of polymer extruded in FDM,are more likely to produce high resolution,non-porous parts suitable for high pressure flow chemistry applications.StereolithographyStereolithography (SL)is a technique that uses layer-by-layer photo-polymerisation of a liquid resin to create solid three-dimensional parts.SL is a more accurate and reproducible manufacturing technique than FDM,demonstrating much higher build resolutions and fully dense parts.Smaller SL machines are capable of producing parts at a layer thickness of 0.05mm and accuracies of ±0.1mm are achievable over smaller parts (less than 100mm for example)and even over larger parts dimensional accuracy is ser spot size can be varied but is typically around 0.2mm;however,the mini-mum wall thickness achievable in practice is around 0.5mm.Although capable of thinner layers,larger SL machines are typically operated at thicker layers (0.1to 0.15mm)for eco-nomic reasons.With a well maintained SL machine repeat-ability is very good.Dimensional accuracy of completed parts can be affected by environmental conditions that the final part is exposed to.Prolonged exposure to UV light and exces-sive humidity can lead to softening and distortion of parts,especially when subject to continued stress (i.e.creep).This is typically avoided by dry storage and sometimes with the use of protective lacquers.In chemistry applications SL has several significant advantages as a manufacturing process,which makes it an excellent choice for prototype reactor manufacture.The superior build resolution allows narrow flow paths and com-plex internal structures to be realised.27The solid part is produced in a bath of the liquid,un-cured resin,making the production of the parts with hollow sections (required for flow)simple,with the resin simply draining clear from the channels.In addition,parts have good optical clarity (although opaque resins are available)allowing internal structures and defects to be easily visualised,and reactions to be visually moni-tored as they proceed.Unfortunately,SL material choices aretypically limited to a narrow range of UV-curable photo-polymers,typically based on acrylates,epoxides and urethanes.Although the final material is likely to be densely cross-linked,conferring better mechanical properties and solvent resistance than many polymers,it is still vulnerable to destructive swell-ing in some solvents.The commercially available SL resin used in our work,3D Systems Accura 60,contains a cycloaliphatic diepoxide,polymerised by a photo-cationic initiator,and an aliphatic tetra-acrylate,polymerised by a photoradical initia-tor.28We found the cured Accura 60material to show mini-mal swelling upon immersion for 24hours in a range of organic solvents,but demonstrated significant swelling in solvents such as THF and dichloromethane.Although this particular polymer is thermally stable to 150°C,the heat deflection temperature is relatively low at around 50°C and the material is poorly conducting.29Above this temperature the polymer becomes less rigid and may not be suitable for high-pressure work.This limits the range of chemistry avail-able when using Accura 60parts to low temperature reac-tions in milder solvent systems.However research is on-going to develop new SL materials with better chemical and mechanical properties,with materials now available which are comparable to commercially used polymers such as ABS,PC and PEEK.Ceramic filled resins are also available which offer excellent mechanical properties and increased resis-tance to solvents,although their comparatively high viscosity means that very narrow channels become difficult to clear.As proof of concept,we manufactured a reactor from Accura 60(RD2)consisting of convoluted tube with length 3300mm,diameter 3mm and a total volume of 23mL,using a 3D Systems Viper si2SLA system.30This reactor was found to be completely leak-free under operating pressures up to 30bars.The materials used were as supplied by the manufacturer and within recommended service life (Fig.2).The use of this reactor for flow chemistry has been dem-onstrated using a simple functional group interconversion reaction –the oxidation of an aldehyde to a methyl ester.Jamison et al.demonstrated this reaction using a FlowSyn system equipped with a polymer reaction tube(PFA,Fig.2CAD drawings of the SL reactor RD2(left and bottom right),RD2during the aldehyde oxidation reaction (top right).Lab on a ChipPaperP u b l i s h e d o n 03 S e p t e m b e r 2013. D o w n l o a d e d b y X i a m e n U n i v e r s i t y o n 17/11/2015 11:46:59.perfluoroalkoxy copolymer),permitting a direct comparison with our work with an additively-manufactured reactor.31Inlet A was charged with a solution of tetra-butyl ammo-nium bromide,methanol and the appropriate aldehyde,whilst inlet B was charged with a solution of dilute sodium hypochlorite (bleach).The combined flow streams were pumped at a flow rate of 0.77ml min −1allowing a residence time of 30minutes.The disappearance of the aldehyde and the subsequent appearance of the ester linkage was con-firmed by 1H NMR and FTIR.Results are shown in Table 1.In nearly all examples,yields and conversions are compa-rable to,or even greater than,those obtained in a conven-tional reactor.The design freedom afforded by AM allows more complex internal reactor geometries to be explored.For example,a reactor comparable in design and dimensions to RD2was manufactured,except with a second inlet incorporated half way through the flow path (RD3).An appropriate reaction was determined as being the two step synthesis of Sudan 1,a red colouring agent used in the textile industry.The reac-tion was demonstrated using flow chemistry in 2002by Wooton et al,32who carried out the reaction using a commer-cially available glass chip micro-reactor.The in situ genera-tion and quenching of the unstable diazonium intermediate,is an ideal reaction to demonstrate the capabilities of this type of reactor design.31Inlet A was charged with a solution of aniline,hydrochloric acid and dimethylformamide (DMF),whilst inlet B was charged with a solution of sodium nitrite and DMF.The combined flow streams were pumped at a flow rate of 0.39ml min −1allowing a residence time of 30minutes.At this point a solution of 2-naphthol and DMF were introduced through inlet C at a flow rate of 0.78ml min −1to give a total residence of 45minutes.The disappear-ance of the amine and the subsequent appearance of the diazo linkage were confirmed by 1H NMR and FTIR.To demonstrate the ability of SL to produce very complex parts capable of withstanding high pressures,another larger split-and-recombine (SAR)mixer was built (RD4).The effi-ciency of the mixer was experimentally determined via a competing parallel reaction.Elemental iodine was generated in the reaction between potassium iodide and potassium persulfate,whilst being simultaneously reduced back to iodide by competitive sodium thiosulfate (Fig.3).Once the thiosulfate is consumed the elemental iodine will complexto a starch buffer,leading to an instantaneous colour change from colourless to dark blue.Mixing efficiency within the system is directly proportional to the time taken for this colour change to occur.The reaction was undertaken in both mixer RD4and tubing of comparative dimensions across a number of flow rates.Inlet A was charged with a solution of potassium iodide,whilst Inlet B was charged with a solution of potassium persulfate,sodium thiosulfate and starch indicator.When comparing the times to colour change between the two systems,it was shown that mixer RD4decreased the reaction time by as much as 17%,dem-onstrating improved mixing efficiency.Other work has also seen a range of tubes (RD5)manufactured,with internal diameters ranging from 0.5–1.5mm.This part demonstrates the capability of SL in manufacturing sub-millimetre reactors,making this technique directly comparable to current com-mercially used micro-reactor manufacturing methods (Fig.4).Table 1The oxidation of a range of aldehydes to their corresponding methyl esters -all reactions carried out in reactor RD2R group Conversion (%)Yield (%)AM reactor Conventional tube reactor (Jamison 31)AM reactor Conventional tube reactor (Jamison 31)Ph93588651p -CH 3C 6H 454565045p -NO 2C 6H 410010099>95p -BrC 6H 463855281m -NO 2C 6H 4991009995PhCH 2CH 2100969841(CH 3)2CH 100N/A 88N/A p -OMe C 6H 45657348Fig.4Comparing mixing efficiency of split-and-recombine mixer RD4(“AM reactor ”)to conventional PTFE tubing.Reaction time until a colour change for an iodide/persulfate/thiosulfate reaction is shown for the two reactors operating at various flowrates.Fig.3Split-and-recombine mixer RD4during the competitive reaction of elemental iodine and potassium iodide.Lab on a ChipPaperP u b l i s h e d o n 03 S e p t e m b e r 2013. D o w n l o a d e d b y X i a m e n U n i v e r s i t y o n 17/11/2015 11:46:59.Multi-jet modelling (MJM)Multi-jet modelling (MJM)was also used to produce a split-and-recombine (SAR)mixer.MJM utilises a large number of very small ink-jet type print heads to deposit liquid photo-polymerising monomers to produce the layers.The layers are instantaneously cured by exposure to UV lights as they are printed.There are a variety of types of machine available.In this work a mixer was made using a 3D Systems ProJet 3000Plus 3D Printer.The materials used were as supplied by the manufacturer in sealed cartridges and within recommended service life.This machine can print in exceptionally thin layers,in this case 16microns,and produces fully dense parts.It also has the advantage of utilising a wax support material that can be melted out from complex internal cavi-ties easily.This enables very narrow,complex channels to be made accurately.The materials are acrylate based and conse-quently the physical properties and solvent resistance is not as good as the most recent hybrid SL resins.However,broadly speaking the materials used are similar to those used in SL and consequently for chemistry applications the advan-tages and disadvantages of this type of process are essentially the same as for SL.The range of complex reactors designed and manufactured in this research demonstrates the applicability of both SL and MJM towards novel milli-scale chemical reactors.The poten-tial for producing highly intricate reactors,coupled with the incorporation of further design complexities such as analyti-cal functionality make both these techniques exciting future areas of research.Although both techniques have a number of desirable manufacturing properties,the currently limited materials choice available for these processes does restrict the chemical applications of this reactor type.We therefore sought AM manufacturing processes which could overcome these material ser sintering (LS)Laser sintering (LS)is a powder-based AM technique that uses a high-powered laser to scan and consolidate powder particles into a solid free-form object.33LS is capable of building high resolution parts from a wide range of powder materials including polymers,34ceramics 35and metals,36however the term LS is usually reserved for polymers.Resolu-tion is limited by a combination of the particle size (typically 50–100μm)and the laser spot size.The polymer materials available for LS are suitable to the demands of a flow reac-tor,with a number of more chemically and thermally com-patible materials available allowing a more diverse range of chemistry to be performed within the reactor.Support struc-tures required in techniques such as extrusion based tech-nologies are not required;as the part being built is supported by the surrounding powder,removing the need for post-processing steps,and consequently removes many limita-tions associated with geometrical design.However,since laser sintering consolidates particles together,porosity of parts may cause leakages under elevated pressures (i.e.powderparticles are not normally fully melted).This porosity is caused by the high viscosity of thermoplastic melts,lead-ing to areas of incomplete particle fusion within the build.The same reactor design used during the SL manufactur-ing process was recreated,using LS to manufacture a reactor from nylon-12(polyamide-12),a well-developed and com-monly used laser sintering material (RD6).It is common practice to use a mixture of fresh powder and recycled powder (i.e.unsintered powder from previous builds)and in this case the powder used was 20%recycled,80%virgin;well within manufacturer recommendations.LS is a well-established AM process and consequently repeatability is good when well-maintained machines are operated by expe-rienced users.Repeatability is also related to the materials processed by LS,hence using Nylon 12,which achieves rela-tively consistent results.Dimensional accuracy is compara-ble to professional standard FDM machines but not quite as good as yer thickness is usually between 0.1and 0.2mm,typically 0.15mm.Typically,dimensional accura-cies of ±0.2mm are achievable on small parts (e.g.below 200mm).The laser spot size is comparatively large com-pared to SL at around 0.6mm which dictates the minimum wall thickness achievable.LS parts are dimensionally stable and comparatively tough but can be subject to distortion over time.Unlike SL where the uncured material was in a liquid form and could be easily drained from channels and voids within the build,the unused LS powder material remained trapped in situ despite attempts to remove it manually by agitation,sonication,compressed air,etc.Where internal channels were long and narrow it became increas-ingly difficult to remove this unused powder.The part was sliced into sections to investigate the internal structures of the build.It was found that the un-sintered powder had formed a dense semi-solid that was difficult to remove or machine out.It is believed that unwanted partial consolida-tion of nylon powder caused by heat conduction from the internal surface of the tube leads to some blockages in the reactor channel,preventing powder removal.Semi-crystalline thermoplastics such as nylon often demonstrate a large degree of shrinkage upon cooling,through both normal thermal shrinkage but also crystallisation processes.Typically this can cause thermoplastics to shrink by as much as a few percent by volume.37This shrinkage will have reduced the internal dimensions of the tube,compressing the powder within.To investigate the minimum tube dimension from which powder could be removed,a series of linear tubes of varying diameter were produced.Again it proved difficult or impossible to remove the powder from any of the tubes,including the larger 5mm diameter tubes.The difficulty in removing powder from the hollow thermoplastic LS parts,presents a fundamental challenge to the use of this tech-nology for reactors.However,it is highly possible that these issues could be reduced through the optimisation of pro-cessing conditions,further material development and inno-vative reactor design,leading to SLS becoming a highly desirable process for reactor manufacture.Lab on a ChipPaperP u b l i s h e d o n 03 S e p t e m b e r 2013. D o w n l o a d e d b y X i a m e n U n i v e r s i t y o n 17/11/2015 11:46:59.Selective laser melting (SLM)SLM is a powder-based AM technique similar to LS,which uses a higher power laser to completely melt metal powder into a single solid body.38Re-melting between layers and the lower viscosity of liquid metal compared to polymer ensures fully dense parts and relatively isotropic physical properties.SLM is capable of manufacturing parts in a range of chemi-cally inert and thermally stable metals such as stainless steel,39titanium 40and aluminium.41In fact one of the commonest materials applied in SLM is 316L Stainless steel,a material frequently used in flow chemistry.SLM can produce fully consolidated parts with minimal porosity,and mechanical properties comparable to bulk metals.42The shrinkage on solidification for metals is typically much less than for semi-crystalline polymers which should facilitate powder removal.SLM type machines are available from a number of commer-cial manufacturers and in a variety of sizes.However,they generally utilise a fibre laser of 100to 400Watts and build in comparatively thin layers,typically 0.02to 0.1mm.It is common to recycle unused powder which is sieved to remove oversized particles.Powders may also be dried to improve flow during the build process.Unlike thermoplastic LS,unused powder remains unaffected and can be re-used.Dimensional accuracy is highly geometry dependent and potentially affected by heat distortion as well as inherent machine characteristics.However,accuracies of ±0.05mm can be achieved over small ser spot size can be between 0.3and 0.7mm which dictates the minimum wall thickness achievable.Repeatability is generally good once a particular part has been optimised for the build process.SLM has been used to manufacture metallic chip reactors with 100μm wide channels,mimicking commercially avail-able glass chip reactors.35However,the resolution of SLM parts is still considerably lower than commercially available equivalents,and at present is not a viable alternative.Indeed the inherent surface roughness associated with the SLM manufacturing process is considerable when dealing with micro-channels of this magnitude (depending on the average particle size used)making it unlikely that any further com-plexity could be incorporated into this design.However,the achievable resolution is good enough to be considered for milli-scale reactor manufacturing.Initially,a simple proof of concept reactor was manufactured from titanium (RD7)using a commercially available Renishaw AM250,43in order to determine the suitability of the process.The reactor consisted of a tube of length 300mm,diameter 3mm and a total volume of 2.1mL.The remaining un-melted powder was successfully removed from the reactor allowing us to demonstrate that the metallic reactor was non-porous and stable under elevated pressure (25bar)and over a range of temperatures.Having successfully manufactured this minia-ture prototype we built a second larger titanium reactor of length 4210mm,diameter 3mm and a total volume of 13.23mL.Unfortunately,the un-melted powder again proved difficult to remove from the narrow internal structures of the build.After the reactor was sliced into segments,it wasnoticeable that some unwanted partial melting of the metal powder had occurred within the tubes,causing significant blockages.We hypothesised that the bulk mass of metal sur-rounding the tubes was leading to considerable heat conduc-tion towards the tubing network,leading to some unwanted partial melting of the metal powder within.It was presumed that by removing the bulk mass of metal from around the tubes,this partial joining effect would be substantially reduced.We experimented by building a number of small titanium tubes,varying the internal dimensions from 2mm to 3mm diameter and the wall thicknesses of each.The powder was successfully removed from each tube,demon-strating that SLM has potential in the manufacture of milli-scale flow reactors.More recent work has used a commercially available Realizer SLM 5044to manufacture a similar range of stainless steel tubes (RD8)with internal diameters ranging from 1to 2mm.The resolution of this build was substan-tially increased from previous work and powder removal proved unproblematic from any of the internal structures.Future work is planned to manufacture a range of more com-plex reactor systems (Fig.5).However there are significant problems to overcome before SLM can be effectively and routinely applied to the manufacture of reactors.The build resolution associated with SLM is relatively poor when compared to conventional manufacturing techniques such as computer-numerically controlled machining (CNC)or electro-discharge machining (EDM),but its wide range of appropriate materials and geo-metric capability offer considerable scope.Whether or not the powder can be removed from more complex geometries is unclear and undoubtedly requires more research,as the potential for SLM as a manufacturing tool for reactor tech-nologies is substantial.Review of AM techniquesThe additive manufacturing techniques presented in this paper each demonstrate significant future scope in terms of chemical reactor design and manufacture.However,each of the mentioned techniques is not without its own engineering and material issues,which need to be fully considered before they could be used to manufacture chemical reactors.TheFig.5Selective laser melted RD8demonstrating the potential build resolution possible using SLM.Lab on a ChipPaperP u b l i s h e d o n 03 S e p t e m b e r 2013. D o w n l o a d e d b y X i a m e n U n i v e r s i t y o n 17/11/2015 11:46:59.。
核安全文化的十项特征
档案INPO12-0122012年12月INPOTraits of a Healthy Nuclear Safety Culture健全的核安全文化特征OPEN DISTRIBUTION 公开发行OPEN DISTRIBUTION: Copyright © 2012 by the Institute of Nuclear Power Operations. Not for sale or commercial use. All other rights reserved.公开发行：版权©2012 美国核电运行研究所。
非卖品。
版权所有。
NOTICE: This information was prepared in connection with work sponsored by the Institute of Nuclear Power Operations (INPO). Neither INPO, INPO members, INPO participants, nor any person acting on behalf of them (a) makes any warranty or representation, expressed or implied, with respect to the accuracy, completeness, or usefulness of the information contained in this document, or that the use of any information, apparatus, method, or process disclosed in this document may not infringe on privately owned rights, or (b) assumes any liabilities with respect to the use of, or for damages resulting from the use of any information, apparatus, method, or process disclosed in this document.注：该信息是由美国核电运行研究所发起准备的。