第32卷第3期 2017年6月
新型炭材料
NEW CARBON MATERIALS
Vol. 32 No. 3
Jun. 2017
文章编号：1007-8827(2017)03鄄0193-12
用于高温气冷堆的核石墨
周湘文，唐亚平，卢振明，张杰，刘兵
(清华大学核能与新能源技术研究院，先进核能技术协同创新中心，先进反应堆工程与安全教育部重点实验室，北京100084)
摘要：自1942年首次在CP-1反应堆中使用以来，核石墨因其优异的综合性能，在核反应堆特别高温气冷堆中被广泛使 用。作为第四代候选堆型之一，高温气冷堆主要包括球床堆和柱状堆两种堆型。在两种堆型
中，石墨主要用作慢化剂、燃料 元件基体材料及堆内结构材料。在反应堆运行中，中子辐照使得石墨的相关性能下降甚至可能失效。原材料及成型方式对 于石墨的结构、性能及其在辐照中的表现起到决定性的作用。辐照中石墨微观结构及尺寸的变化是其宏观热力学性能变化 的内在原因，辐照温度及剂量对于石墨的结构及性能变化起决定性作用。本文介绍了高温气冷堆中核石墨的性能要求及核 石墨的生产流程，阐述了不同温度及辐照条件下石墨热力学性能及微观结构的变化规律，并对当前国内外核石墨的研究现状 及未来核石墨的长期发展如焦炭的稳定供应和石墨的回收进行讨论。本文可为有志于研发用于未来我国商业化的高温气冷 堆中的核石墨的生产厂家提供参考。
关键词：核石墨；高温气冷堆；辐照；微观结构；物理、力学及热学性能
中图分类号：TQ127.1 + 1文献标识码：A
基金项目：国家公派留学基金（201406215002)；国家科技重大专项（ZX06901)；清华大学自主科研项目（20121088038).
通讯作者：周湘文，副教授，博士. E-mail: xiangwen@ tsinghua.edu. cn
Nuclear graphite for high temperature gas-cooled reactors
ZHOU Xiang-wen，TANG Ya-ping，LU Zhen-ming，ZHANG Jie，LIU Bing (Institute o f Nuclear and
New Energy Technology o f Tsinghua University，Collaborative Innovation Center o f Advanced Nuclear Energy Technology，the key laboratory o f advanced reactor engineering and safety，Ministry o f Education，Beijing100084，China)
Abstract: Since its first successful use in the CP-1 nuclear reactor in 1942，nuclear graphite has played an important role in nucle­ar reactors especially the high temperature gas-cooled type (HTGRs) owing to its outstanding comprehensive nuclear properties. As the most promising candidate for generation IV reactors，HTGRs have two main designs，the pebble bed reactor and the prismatic re­actor. In both designs，the graphite acts as the moderator，fuel matrix，and a major core structural component. However，the me­chanical and thermal properties of graphite are generally reduced by the high fluences of neutron irradiation of during reactor opera- tion，making graphite more susceptible to failure after a significant neutron dose. Since the starting raw materials such as the cokes and the subsequent forming method play a critical role in determining the structure and corresponding properties and performance of graphite under irradiation，the judicious selection of high-purity raw materials，forming method，graphitization temperature and any halogen purification are required to obtain the desired properties such as the purity and isotropy. The microstructural and correspond­ing dimensional changes under irradiation are the underlying mechanism for the changes of most thermal
and mechanical properties of graphite，and irradiation temperature and neutron fluence play key roles in determining the microstructural and property changes of the graphite. In this paper，the basic requirements of nuclear graphite as a moderator for HTGRs and its manufacturing process are presented. In addition，changes in the mechanical and thermal properties of graphite at different temperatures and under different neutron fluences are elaborated. Furthermore，the current status of nuclear graphite development in China and abroad is discussed，and long-term problems regarding nuclear graphite such as the sustainable and stable supply of cokes as well as the recycling of used material are discussed. This paper is intended to act as a reference for graphite providers who are interested in developing nuclear graphite for potential applications in future commercial Chinese HTGRs.
Key words：Nuclear graphite；High temperature gas-cooled reactors；Irradiation；Microstructure；Physical，mechanical and ther­mal properties
Received date：2017-02-26；Revised date:2017-05-13
reactor 性能Foundation item：State Scholarship Foundation of China (201406215002) ；Chinese National S&T Major Project (ZX06901) ；Tsin­ghua University Initiative Scientific Research Program (20121088038).
Corresponding author：ZHOU Xiang-wen，Associate Professor. E-mail: xiangwen@ tsinghua. edu. cn
English edition available online ScienceDirect ( www. sciencedirect. com/science/journal/18725805 ).
DOI：10. 1016/S1872-5805(17)60116-1
• 194•新型炭材料第32卷
1Introduction
The phrase nuclear graphite began to be used at the end of 1942 when the first nuclear fission occurred in the graphite moderated nuclear reactor CP-1[I]. From the early 1960s, the United Kingdom, the Unit­ed States and Germany began to develop high temper­ature gas-cooled reactors (HTGRs). Japan began the construction of a 30 MWth high temperature test reac­tor (HTTR) in 1991, which reached its first criticali­ty in 1998. In China, a 10 MW experimental high temperature gas-cooled reactor ( HTR-10 )[23], whose design started in 1992 and construction com­menced in 1995, reached it criticality in the end of 2000, and its full power in the beginning of 2003. Since the Fukushima accident in March, 2011, the public has paid more and more attention to the safety of nuclear power. As a candi
date reactor for the Gen- eration-IV reactors, the construction of a 2x250 MW high temperature gas-cooled reactor pebble-bed mod­ule (HTR-PM) with inherent safety is underway in Shidao Bay, Rongcheng of Shandong province, Chi­na and is expected to complete in 2017[4]. In both of the research and commercial HTGRs, the reactor re­flectors and cores have been constructed by structural graphite components. Past designs represent two pri­mary core concepts commercially favored for HTGRs ：the prismatic block reactor (PM R) and the pebble- bed reactor (PB R)[2]. In both of the HTGR concepts the polycrystalline graphite not only is a major struc­tural component which offers thermal and neutron shielding and provides channels for fuel and coolant gas, channels for control and safety shut off devices, but also acts as a moderator and matrix material for the fuel elements and control rods and a heat sink or conduction path during reactor trips and transients.
The polycrystalline graphite exhibits significant importance in HTGRs because of its outstanding nu­clear physical properties such as high moderating and reflecting efficiency, a relatively low atomic mass and a low absorption cross-section for neutrons, in addi­tion to high mechanical strength, good chemical sta­bility and thermal shock resistance, high machinabili- ty and light weight[5]. The following example illus­trates the importance of nuclear graphite in more de­tails. For the thorium high temperature reactor ( TH- TR) in Germany with a power of 300 MWe, nearly 400 000 kg of nuclear gra
phite has been used[2] •In China, approximately 60 tons of graphite was used in HTR-10[3], and more than 1000 tons of nuclear graphite will be used in HTR-PM as the structural ma­terial and matrix graphite of pebble fuel elements ⑷. The raw materials of matrix graphite of fuel elements for HTR-10 and HTR-PM such as natural flake graph­ite and artificial graphite powder are supplied by Chi­nese domestic providers[6,7]. The behavior of the in­dividual fuel particles and the matrix graphite material in which the particles are encased are not considered here. However, it should be noted that although the graphite technology associated with the matrix graph­ite is related to that of the main structural graphite such as the moderator there are differences as non- graphitized materials and natural flake graphite are used in the matrix graphite. Because so far no quali­fied domestic nuclear graphite is available, all the structural nuclear graphite materials for HTR-10 and HTR-PM are imported from Toyo Tanso of Japan. In April 2015, China Nuclear Engineering Corporation Ltd ( CNEC) announced that its proposal for two commercial 600 MWe HTGRs (HTR-600) at Ruijin city in Jiangxi Province had passed an initial feasibili­ty review. The HTR-600 is planned to start construc­tion in 2017 and for grid connection in 2021[8]. In or­der to achieve the economy and security of supply, the structural nuclear graphite must be provided by domestic providers in China in the future. Fortunate­ly, with the rocketing development of photovoltaic in­dustry in China, several Chinese companies have emerged which can produce the fine-grained isotrop­ic, isostatic molded, high strength graphite in large scale. Some of the
manufacturers with state-of-the-art graphite manufacture capabilities should be chosen as the potential candidate providers of the structural nu­clear graphite for HTGRs based on qualification pro­grams. However, during the operation of a reactor, many of the graphite physical properties are signifi­cantly changed due to the high fast neutron doses. The physical, mechanical and chemical properties of graphite can be influenced negatively by irradiation induced damage, which would lead to the failure of graphite components. In pebble-bed HTGRs such as HTR-PM in China, the core support graphite structure is particularly considered permanent, although it is expected that certain high neutron dose components ( inner graphite reflector) will be replaced during the whole lifetime of the reactor. During the life time of the reactor, the reflector graphite would be subjected to a very high integrated fluence of fast neutrons of around 3x1022n/cm2(E>0.1M eV)[910]. Therefore, the pre-irradiation and post-irradiation comprehensive properties of nuclear graphite candidates must be thor­oughly examined and evaluated. Those properties of nuclear graphite are strongly dependent on the extent of anisotropy, grain size, microstructural orientation and defects, purity, and fabrication method.
In this paper, basic nuclear requirements of nu­
第3期ZHOU Xiang-wen et a l：Nuclear graphite for high temperature gas-cooled reactors•195.
clear graphite are presented and the specifications such as the manufacture, material properties with three pri­mary areas (physical, thermal and mechanical) and irradiation responses of nuclear graphite suitable for HTGRs are elaborated, which could be a reference for the potential providers who are anxious to develop the nuclear graphite for future commercial HTGRs of Chi­na. The long-term considerations such as those invol­ving the cokes and recycle for nuclear graphite are al­so discussed.
2 Nuclear requirements of graphite for HTGRs
2.1 Fission reactions with neutrons
The tremendous energy produced in HTGRs is from the fission of isotopes such as 92 U233，92 U235，and 94Pu239 . Fission of a heavy element，with release of energy and further neutrons，is usually initiated by an impinging neutron. The fission of 92U235 can be de­scribed as：
92『5+。^寅A+B + v * n+ 〜200 MeV ⑴
The average yield per fission of 92 U235is about 2.5 fast neutrons. The energy of neutrons released from the fission reactions can be described by a Max­wellian distribution，with an average value of approx­imately 2 MeV. The probabilities of the fission reac­tions initiated by neutrons ( the cross section)
are in­versely proportional to the velocity of the neutrons. It is essential to slow down the “fast” neutrons yielded by fission to “thermal” neutrons with lower energies (〜0. 025 eV at room temperature)，which corre­spond to a neutron velocity of 2. 2x103 ^s.The slo­wing down process results principally from energy transfer during elastic collisions between the neutrons and medium which is commonly called “ moderation” and the non-absorbing medium where the moderation takes place outside the fuel is termed “ moderator” . As is known，the nuclear fuel for HTGRs is common­ly a mixture of low-enriched 92U235 and 92U2'Once the moderated thermal neutrons return to the fuel，they are most likely to cause fission in the 92 U235，in­stead of being captured by 92 U238 .
2.2 Nuclear requirements for a good moderator
There are two very fundamental nuclear require­ments for any moderator in HTGRs. First，it must have a small cross section for neutron absorption. Second，fast neutrons must be effectively slowed down to thermal neutrons over short distances and within few collisions in the moderator. Thus，the probability of neutron absorbed by the moderator im­purities and，92U238or absorbing structural materials in a reactor is reduced. Therefore，a good moderator material should exhibit a high slowing down power and low adsorption ability[n，l2]. Furthermore，the moderator material should be economically acceptable and compatible with the other materials used in the core of reactor，and mai
ntain physical and chemical stability against the bombarding neutrons.
According to the fundamental consideration of Newton 爷 s law，the more energy loss of neutron per collision takes place when the target nuclei have the lower atomic mass. The average logarithmic energy change per scattering，孜，can be described as equation (2)：
孜  2 (2)
A+2/3
Where，A is the atomic mass of the collision nu­cleus. Therefore，the moderator should be based on the elements of low atomic mass，which actually lim­its the choice to elements of atomic number (Z) less than sixteen. The parameter 孜is a good index of the moderating ability of a material，but it is also depend­ent on the chances of a scattering collision occurring (scattering cross section，撞s ). The slowing down power (SDP)，孜* 撞s，which takes the two aspects a­bove into consideration is frequently used to indicate the ability of a moderator material to slow down neu­trons. However，since the parameter SDP is inde­pendent of the neutron absorption property of a mate- rial，it alone is not enough to guarantee a particular material to be a good moderator. Therefore，the ratio of the SDP to the absorbing cross section (撞a)，which is referred to as the moderating ratio，should be a better ind
ication to evaluate a particular material as a moderator. A shortcoming of the moderating ratio is that it does not take the density of a material into con­sideration. For example，some gases such as helium have a high value for the moderating ratio，but they are of little use as moderators because of their low density at normal pressures and temperatures. Accord­ing to the requirements of a good moderator men­tioned above，the choice of potential moderator candi­dates of practical use thus reduces to the four materials listed in Table 1.
Table 1 Moderating properties of candidate
moderator materials[13].
Moderator
materials
S D P(孜* 撞s)
(cm-1)
Moderating
ratio
Density
(g • cm-3)
h2o  1.53072  1.000
d2o0.37012000  1.106
Be0. 176159  1.852 Graphite0.064170  1.70-1.85
• 196•新型炭材料第32卷
The materials in Table 1are either in liquid or solid states, which present acceptable density proper­ties. It can be seen that the deuterium oxide (heavy water) has the highest value of moderating ratio. But the costs to separate the heavy water from the ordinary water are high and the heavy water is likely to leak out. The initial capital cost and leakage make_up cost are too high to make the heavy water as the first choice for the moderator. Ordinary water is very easy to obtain with low cost and relatively unaffected by neutron irradiation. However, neutron absorption by hydrogen reduces the moderating ra
tio. As a result, enriched uranium fuel should be used in water moder­ated reactor to achieve the required neutron economy. Beryllium is not a good moderator for HTGRs because it is expensive, hard to machine, and highly toxic. Finally, graphite is a good moderator, which offers an acceptable compromise between nuclear properties, cost and utility as a structural material for the core of HTGRs[l3]. However, the use of graphite in HTGRs as a moderator has its problems. The significant effects of neutron irradiation dose on the structure and corresponding properties of graphite should be ad­dressed and will be presented later.
3 Nuclear graphite under high tempera­ture and irradiation[5,13]
The nuclear graphite has been proved to be a via­ble structural material for HTGRs according to the de­velopment of HTGRs. The specially developed grades of nuclear graphite can meet the design requirements of HTGRs. Characteristics required for modern nucle­ar graphite are ：1) high purity with a low elemental contamination (especially boron )；2) acceptable iso­tropic or near isotropic dimensional change ；3) ma- chinability into large graphite components ；4) valida­ted irradiation database. While these are fundamental and necessary attributes to obtain acceptable compo­nent lifetimes for use within a HTGR irradiation envi­ronment, they may not be sufficient to demonstrate adequate structural integrity for all design configura­tions. Some typical property requirements of struct
ur­al nuclear graphite for HTR-PM are shown in Table 2.
Table 2 Typical property requirements of nuclear graphite for HTR-PM.
Property Unit Extrusion Vibration molded Isostatic molded
Grain size mm臆1.50臆1.00臆0.04
Density g/cm3逸1.75逸1.75逸1.76 Thermal conductivity *W/(m •K)逸125逸125逸125 CTE *10-6/K臆4.5臆4.5臆4.0 Anisotropic factor-臆1. 10臆1.05臆1.04
Tensile strength*MPa逸20.0逸20.0逸25.0
Compressive strength*MPa逸65.0逸65.0逸75.0
Boron equivalent content ppm臆0.90臆0.90臆0.90 Ash content ppm臆100臆100臆100 *：The properties in two directions of with grain and against grain must be provided.
So far the basic mechanisms of irradiation behav­ior of nuclear graphite are well understood. Individual grades of nuclear graphite will exhibit distinctly differ­ent responses to the irradiated environments base
d on the degree of anisotropy, purity, grain size, micro­structure orientation, microstructural defects and forming method. However, how these in-crystal effects exactly interact with the structure of bulk nu­clear graphite and its subsequent dimensional and property changes are not fully understood. While the behavior of any given graphite can be predicted in broad terms, the exact magnitude of irradiation in­duced changes cannot yet be precisely predicted with models based on previous historical data. Because of its unique structure and texture, each grade of graph­ite may exhibit different irradiation behaviors to some degree.
The texture introduced during forming and ther­mal processes impart property variations within a billet of nuclear graphite. There are also property differ­ences in the forming direction compared with the per­pendicular to the forming direction. Moreover, a den­sity gradient should exist from billet center to edge. These variations must be quantified for any given grade of nuclear graphite. In addition, there are sta­tistical variations in properties between billet to billet within a batch, and between production batches be­cause of variations in raw materials, formulations and processing conditions. The inherent variability to as- produced graphite must be assessed through the analy­sis of large sample populations to determine the maxi­mum range of material property variations expected for components machined from an average billet. The larger the sample population in each billet is ana­
第3期ZHOU Xiang-wen et a l：Nuclear graphite for high temperature gas-cooled reactors•197-
lyzed, the higher the resolution of the distribution of the pertinent properties.
Because the material properties of the graphite will change or degrade during reactor environment ex­posure, the newly fabricated grades of nuclear graph­ite must be characterized and irradiated to demonstrate that current grades of graphite exhibit acceptable pre­irradiated and irradiated properties so that the thermo­mechanical design of the structural graphite in the HT- GRs can be validated. The property changes or degra­dations are dependent upon several factors, such as temperature, neutron fluence, graphite microstruc­ture ,purity and applied stresses during operation, and the temperature and neutron fluence play the key role in determining the microstructure changes and the property responses of the graphite. Material property values with three primary areas such as physical (mi­crostructure ),mechanical and thermal properties should be presented and discussed in the following sections.
3.1 Physical properties
The physical changes related to the graphite mi­crostructure induced by irradiation exposure are the potential mechanism for most physical, mechanical and thermal issues in the graphite components. In
common, physical properties of graphite are related to the characteristics of its microstructure.
3.1.1 Microstructure
In graphite, parallel hexagonal carbon layers where the atoms are covalently bonded (524 kJ/mol) are stacked by a much weaker van der Waals bond (7 kJ/m ol). Under neutron irradiation, the carbon atoms are knocked out from their equilibrium lattice positions into interstitial positions throughout the mi­crostructure. As a result, the basal planes shrink or collapse as further damage accumulates and vacancy clusters grow within the basal planes. Due to the ani­sotropic crystal structure of graphite, the knocked out atoms preferentially diffuse and accumulate in the lower energy areas between the basal planes. These small clusters of interstitial atoms aggregate into large clusters and finally rearrange themselves into new bas­al planes, leading to the expansion of the graphite crystal in the c-axis direction (perpendicular to the basal planes) and contraction in the a-axis direction ( parallel to the basal planes ) . The microstructures such as the crystalline and grain size, degree of graphitization, anisotropy, variation of crystallite ori­entation ,porosity and micro-damage within the formed graphitic microstructure during fabrication processes of nuclear graphite are the main parameters that affect its macroscopic physical, thermal and me­chanical performances in HTGRs.3.1.2Dimensional changes
As mentioned above, during irradiation, the crystallites of graphite swell in the c-axis direction and contract in the a-axis direction due to the anisotropic crystallite structure of graphite. Processing defects (Mrozowski cracks)[ l4] formed during manufacture cool-down which are preferentially aligned in the crystallographic a-axis direction provide a ready vol­ume of space to physically accommodate the c-axis swelling during irradiation. Since the c-axis swelling is accommodated, the graphite thus undergoes net volume shrinkage because of the a-axis contraction throughout the graphite volume. With further irradia­tion, more and more cracks close and finally the c-ax- is swelling is no longer accommodated. Meanwhile, the incompatibility of crystallite dimensional changes leads to the generation of new porosity oriented paral­lel to the basal planes. As a result, the volume con­traction rate falls and eventually reaches zero. When this reversal or turnaround occurs, the graphite begins to swell at an increasing rate with further irradiation due to the combined effect of c-axis swelling and new porosity or cracks generation. As shown in Fig. 1, increasing the irradiation temperature causes faster crack closure because of thermal expansion of the crystallites. During operation of HTGR, local differ­ences in neutron fluence and temperature within large graphite components lead to differential strains and re­sultant stresses to develop in the graphite microstruc­ture, which should be large enough to induce crack growth and eventual failure of the component within a short in-reactor time. Fortunately, premature failure is avoided by strain relief of induced stresses ( irradia­tion creep) within irradiated gra
phite microstructures, which allows the graphite to withstand irradiation damage resulting from irradiation induced dimensional changes. Typically, the useful lifetime for graphite is defined as the time or neutron fluence it takes for the material to shrink and then swell back to its original dimension. The magnitude and rate of dimensional change and the point of turnaround are controlled by a number of factors, including the degree of crystallini­ty within the microstructure, process conditions, vari­ation of crystallite orientation and the resident micro­damage within the graphite. In addition, the rate of dimensional change is also remarkably affected by the irradiation temperature. Increasing the irradiation tem­perature leads to faster crack closure due to thermal expansion of the crystallites, which results in faster turnaround rates. Consequently, it is necessary to de­termine the dimensional and corresponding volumetric changes as a function of irradiation dose and tempera­ture so that the critical performance evaluations such

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。