X-ray Clusters at High Redshift

phoenix|x-raySystem + Services GmbH quality|assurance version 3.2中文操作手冊规格变动不预先通知图文仅供参考ContentsThe quality|assurance program (5)Installation (5)How to start the quality|assurance program (5)Window Layout ................. (6)The Menu (7)Program (7)Image (7)Edit (7)Setup (8)Info (8)Controls at the upper windows margin (9)Program Load (9)Program Run (9)Image Load (9)Image Save as (9)Print Image (9)Single View (9)Quad view (9)Restore (9)Invert (9)Zoom (9)Text (9)Control at the right window margin (10)F2 Live (10)Crosshair (10)Annotations (10)Shading correction (10)F3 Integrate (10)F4 Grab (10)Contrast Auto (10)Contrast Manual (10)Contrast Equalize (11)Sharpen Weak (11)Sharpen Strong (11)Sharpen Relief (11)Smooth Weak (11)Smooth Strong (11)Wire Sweep (11)Measurement (11)ML-Modul (11)VC-Modul ................................................................................................................... .11 BGA-Modul ................................................................................................................. .11 Controls at the bottom window margin (12)X-ray control (X-ray) (12)Axis-control (CNC) (12)Various Commands (13)Special mouse functions (13)shift image (13)Activate context menu (13)Adjust the size of an image / view (13)Move to clicked position/Zoom area (13)Key board commands (13)Moving the manipulator with the cursor keys (13)Additional print data-Dialog (14)Manual Contrast Adjustment Dialog (15)Digitizer Setup (16)Annotations (17)Types of annotations (17)Add annotation (17)Move annotation (17)Resize annotation (17)Change annotation (18)General (18)Text (18)Lines (18)Recetangle, Ellipse, Arrows (18)Apply annotations to the image (18)Delete all annotations (19)Save annotations (19)Wire Sweep (20)How to operate Wire Sweep (20)Wire Sweep Setup (21)Measurement (22)Conducing a Measurement (22)2 Point Measurement ......,,.. (23)3 Point Measurement (23)CNC-Measurement (23)Measurement Setup (24)Measurement Adjustment (25)Pixel size configuration (25)Load Position Setup (26)Quality assurance Setup (27)ML-Module (28)Performing a ML measurement (29)Measure the distance of two points (29)Measure the distance of two parallel lines (29)Pad measurement and residual ring test (29)Measurement pad offset and residual ring width (30)Adjust pad measuring (30)Adjust (30)The result list (31)Delete and copy results (31)VC-Module (32)Run Voiding Calculation (33)Load VC-Setup (33)Save VC-Setup (33)Save VC-Setup as ................................................................................................................ .33 VC-Setup (33)VC Diagnose (33)Perform Voiding Calculation .................................................................................................. .34 VC-Module Setup .. (35)Pass/Fail Tab (35)Calculation Tab (36)BGA Module (38)Run BGA Check (39)Load BGA-Setup (39)Save BGA-Setup (39)Save BGA-Setup as (39)BGA-Setup (39)BGA Diagnose (39)Running the BGA Analysis (40)BGA Module Setup (41)Pass/Fail Tab (41)Calculation Tab (43)Setup missing balls (45)Teachmode tab (46)Inspection Program (47)Inspection Program Configuration/Modification (48)Program Execution (50)The Inspection Report (51)The INI-File XTS.INI (52)The INI-File CNC.INI (54)The INI-File XRAY.INI (55)The quality|assurance programPhoenix|x-ray 的quality|assurance程序是一项整合CNC和X-Ray控制以提升影像质量的软件,并能符合于生产线,质量保证与实验室中的X-Ray 检查需求.所有的程序结构是依据直觉与容易操作而设计的.程序的功能不仅容易并且能快速显示出X-Ray影像的记录.重要的特征也会特别强调出来.样品可藉由程控来移动和检查,步骤完成迅速,可以储存在程序中, 也可以依需要经常的自动重复.用这种方法,小的、中的、大的样品皆可有效地、容易地检查,这种简易且依觉的检查软件就算对小型样板也一样能支持程序检查.安装如果先前安装的Matrox Image Products 程序仍然存在,则INI档案中的quality|assurance必须储存,而“Matrox ImageProducts” 就必须移除.将安装磁盘“quality|assurance set up”放入CD-ROM中,如果设定无法自动启动“Window Explorer”, 选择D 槽: (假定 D 槽是你的CD-ROM 槽), 请点选“Setup.bat”两次.安装程序包括 3 个步骤:1.Falcon2.Matrox: 有VC或BGA模式,即MIL6.x 和MILLite3.quality|assurance在每个步骤之后,点选window 控制台(黑色背景)并按任何键,当所有的安装步骤完成,请关闭程序并重新启动计算机(开始关闭重新启动计算机)如何启动quality|assurance程序请注意:在启动quality|assurance程序之前,系统的门必须是关闭的,否则连接到X-Ray,CNC系统装置无法被设立.在此状况,先关闭系统的门然后选择function Setup 启动机器或重新启动程序.产生X-Ray幅射,首先X-Ray必须先暖机,在冷机状态下,可经由打开X-Ray开关完成.X-Ray控制系统自动暖机须视机器停机的时间而定,程序需花费15分钟到2小时.在使用控制前,CNC必须先试机完成复归动作,为了在设定检验程序时找到正确位置,在复归时驱动轴必须找到原点的位置.若使用操作杆,操作可以不经试机即可移动CNC位置.当启动程序(系统门关闭),如果需要时你会被告知是否CNC现在需要完成试机和是否X-Ray必须暖机.Window LayoutPhoenix | x-ray 的程序quality|assurance一般都显示在影像文件名称旁的抬头列中. 在抬头列的下方为功能选择单的位置,可点选进入“Program”, “File”, “Edit”和“Info” . 在此选择单下方为“Program files”,” image files”, “display” 和“image zoom” 的按钮组.影像撷取与影像处理的按钮一同在右边的边缘. 在窗口的下方提供实时信息的状态列您可找到X-ray 和CNC控制.在窗口的中央为影像显示的区域,有1到4个影像可供显示,在下方实时信息状态列可显示目前Live影像的区域数.影像用绿色外框强调并且所有的功能只有在此影像可作用.例如,如果您撷取一个影像, 此影像会出现在此活动窗口中,或者,如果您完成一项反差对比,这些只会在此活动窗口有作用. 而此窗口的活动可由移动鼠标的光标或按鼠标的左/右键来控制.红色外框的影像表示此影像为实况影像, 换言之,X-ray 影像显示中(F2 live).在下方实时信息状态列, 对X-ray功能来说,幅射可以启动或关闭,而X-ray控制的电流状态也可以读取.同样Tube的电压数(kV)和电流数(uA)也可以被读取与调整.在下方实时信息状态列, 对CNC功能来说, 可以停止CNC轴的移动,可以读取CNC实时状态, 轴的位置可以读取与控制,操作者也可以移动下载/不下载的位置.下方的状态列提供各种功能的一些协助. 此状态列的左边,依据影像的调整和灰阶指示器,会显示X-ray 和CNC的错误状态. 在右边如果鼠标光标落在一个影像的上面,此影像的调整和相关灰阶值会显示出来. 活动窗口的号码由“View n”提供, n= 1..4. 您也可以读取电流图素的尺寸, 单位,倍率和设定标的物的厚度.功能选择清单在抬头列下方, 窗口的上方有一个功能选择清单提供一般使用的功能.大部份的功能都能由使用这些功能进入.ProgramNew 开启一个新的,空白的程序.Load 由硬盘或磁盘中下载一个程序.Save 储存一个修改过的程序在硬盘或磁盘中.此功能只有在程序修改过之后才可以执行.Save as… 更换新文件名储存程序在硬盘或磁盘中.可用于复制程序时.Edit 编辑“Edit Program”对话来更改程序的调整(请见检查程序结构/修改)Run 执行程序, 启动连续检查(请见程序执行)Reports 能将所有的检验报告记录列在窗口中. 用鼠标右键点选一项记录,拥有open, print, delete等功能context 窗体即会开启Exit 离开程序ImageLoad 由硬盘或磁盘片中下载影像到活动窗口中(绿色外框)Save as…. 储存活动窗口的影像(外框为绿色的)Print 打印所有在显示区的影像(不限为活动窗口的影像)打印与其它有关打印调整的选项都在print-setup的对话盒中1..4 这里是四个最后被下载或储存的影像的文件名.选择其中一个文件名其相关影像会被下载到活动窗口中.Save as Shading Correction image储存实时影像来作明暗修正(请见Shading correction明暗修正)EditCopy 对活动影像作黏贴指令, 黏贴指令会将此窗口影像复制目前时活动窗口中.Paste contentof View n 用copy的指令来复制影像到实时活动窗口中.SetupPrint Setuup启动print-setup的对话盒进入打印机调整打印影像,例如,打印机中每页尺寸,编排,相片,美化Digitizer Setup调整测试板x-ray影像结构的数字化.(请见Digitizer Setup)Wire Sweep 调整wire sweep的最大值(请见wire sweep) Measurement 打开此对话框来调整量测值(请见Measurement Setup)Initialize system确认系统的CNC和X-ray控制连结.如果必要的话,使用者可能会被告知,真空管的暖机与CNC的试机是否应该启动. Load position 用来定义下载的位置.在最后一组程序完成后或当“To load postion”功能被启动,此系统会自动回复到定义下载的位置.BGA Module 调整BGA模式的设定(请见BGA Module设定) quality assurance 调整程序设定(请见quality assurance Setup) InfoAbout quality|assurance显示关于程序的信息Test image 1-3 显示影像来测试监示器和/或打印机上方窗口工具列移动鼠标光标到这些按钮上并按下鼠标左键来完成相对应的功能点选.Program Load由硬盘或磁盘中下载程序Program Run程序执行,启动连续检验(请见Program Execution).Image Load由硬盘或磁盘中下载影像.由此档案选择盒中您可以选择您所要的档案来下载.Image Save As…储存活动窗口(绿色外框)的影像到硬盘或磁盘中.由此挡案选择盒中您可以进入您所要的影像檔名.Print Image打印所有显示区中的影像(不限活动影像).打印与其它有关打印调整的选项都在print-setup 的对话盒中.Single View启动Single View模式.被启动的影像会刚刚好显示在整个显示区中.Quadview启动Quadview模式.所有4个影像会显示Zoom ½ 倍率下.Restore恢复显示.窗口的布置(位置,尺寸)会重新设定到原来的状态下.Invert倒转活动窗口的替代图像.只有影像的替代图像会倒转而并非影像本身.这表示,如果影像是被储存的,此储存影像将不会被倒转.Zoom显示影像的倍率减少与增加分为1/2倍,1倍和2倍Text在影像中加入批注(见Annotations)窗口右方工具列移动鼠标光标到这些按钮上并按下鼠标左键来完成相对应的功能点选或直接按功能键(F2..F4)注意所有的功能只能在活动窗口(绿色外框)中作用.Image:F2 Live显示出实况影像于活动窗口中.可从屏幕中看到移动的样品,此时X-ray影像会不断的出现Crosshair如果选择,crosshair会显示于实况影像上.Annotations如果选择,批注即会出现否则会隐藏.Shading correction对实况影像或已撷取影像完成明暗修正.如果选择单一影像,明暗修正可经由关闭开启完成.明暗修正可排除经由影像系统所造成的灰阶失真.ShadingMode=0(请见INI-File XTS.INI)正常来说此影像是没有标的物的而使用一半x-ray电力(一半电流)和ShadingMode=1(灰阶值将近255)即所谓的明亮影像来作明暗修正.(请见menu Image)F3 Integration完成影像整合.16个影像(系统允许值)会被记录并加总.结果也会分割为16个.此种方式所完成的影像杂质最低.F4 Shapshot在此,由比对到整合只有一个影像会被储存.(Snapshot).完成无杂质缩版.Contrast:Contrast Auto提高自动比对.灰阶影像会自动扩散,进入影像的灰阶值也能以线性方式显示(0..255=黑..白)当实时影像连续自动比对到适当的影像.按Auto键两下将自动提高比对功能关闭Contrast Manual当Auto被定义为自动扩散.Contrast Manual就表示灰阶值范围能以手动方式扩散由0..255 (请见Manual Contrast 调整对话框)Contrast Equalize完成所谓的灰阶值均等.灰阶值的影像为自动扩散,如此每个灰阶值即为相同频动.对于难以比对的影像来说这是非常有用的操作方式.显示为实时影像时,灰阶均等会连续的比对到影像上.按Equalize两下鼠标将均等灰阶值比对功能关闭Sharpen :Sharpen Weak增加影像清析度. 灰阶值的差异度(敏锐度)会被标明.Sharpen Strong与Sharpen Weak功能比较,此影像的清析度效果较强.Sharpen Relief这是一个用左上方光源模拟照明所得的模拟3D的结果. 平面分配128度的灰阶值(中度灰阶),增加灰阶值会显示较明亮画面反之则暗一些. 此功能就算是一个很小的灰阶值也能显示出差异.Smooth:Smooth Weak减少杂质的影像. 如果影像呈现出相当高的杂质,可用此功能来减少杂质. 需注意的是影像的清析度同样会被减弱. 为避免此情况,撷取时应用F3 Integration 来代替F4 Snapshot.Smooth Strong相对于Smooth Weak, 此功能会使杂质与影像清析度一同增强.Wire Sweep完成Wire Sweep量测.(请如何操作Wire Sweep)Measurement完成量测(请参考Measurement)ML-Modul此为选配项目(请参考ML-Module)VC-Modul此为选配项目(请参考VC-Module)BGA-Modul此为选配项目(请参考BGA Module)窗口底部的控制列在窗口底部为X-ray 控制(X-ray)和axis控制(CNC)功能对X-ray功能来说,幅射产生可经由按钮启动与关闭而X-ray控制的电流状态可被读取.而Tube的电压值(kV)与电流数(uA)也可以读取与调整.要调整此数值,OPERATE按钮必需转到“Remote”.对CNC功能来说,可以停止CNC轴的移动,可以读取CNC实时状态, 轴的位置可以读取与控制.操作者也可以移动到下载/非下载的位置.X-ray-control(X-ray):X-Ray On启动x-ray.X-Ray Off关闭x-raynot connected x-ray控制的电流状态未连接.130kV高电压值,点选此钮可调整电压值到其它电压值.300uA实时电流值,点选此钮可调整电流值到其它电流值.Axis-control (CNC):CNC Stop停止所有CNC轴改用操作杆操作.Position reached CNC 实时状态,到达所设定的位置X+160.000X轴的实时位置. 点选此钮可调整X轴到其它位置.关闭x-ray并移动到下载/非下载位置.各种指令:特殊鼠标功能:活动影像/窗口在此影像/窗口,按下鼠标(鼠标的左键或右键)来启动它. 此影像/窗口将会有绿色的外框启动单一画面和四格画面用鼠标的左键按画面两下,来启动单一画面或四格画面移动影像在画面上按下鼠标左键不动,shift key会出现在画面上,移动鼠标光标,影像即可移动到您想要位置.启动context menu在画面上按一下鼠标右键不动,即出现所谓的“对话清单”, 这是一个常用的功能可以直接点选所需功能.调整影像/窗口尺寸移动鼠标光标到影像的外框上或四格画面的中央交叉点上,按下压住鼠标左键,即可自由设定影像尺寸.移动到点选位置/屏幕放缩区在实时影像上(非批注)按下鼠标,此画面会移动到屏幕中央,正常来说,你也必须要按住“Ctrl”键(请见INI-File XTS.INI). 如果你设定一个矩形,此影像可依照矩形的尺寸放缩.(有可能X-ray头真空管会碰撞零件!!)键盘指令:Key 指令简短描述Alt+L Load 下载影像Alt+A Save as… 储存影像到Alt+P Print 打印影像Alt+Q Quad 点择单一画面或四格画面Alt+R Restore 恢复窗口布置Alt+/ Zoom ½ 尺寸减半Alt+1 Zoom 1 正常尺寸Alt+2 Zoom 2 两倍尺寸Alt+S Setup 启动设定F2 Live 启动实时影像F3 Integrate 撷取整合影像F4 Grab 撷取非整合影像, 快照Space Overlay on/off overlay 开启关闭用上下左右键来移动操纵器按住Ctrl 或Shift 键并同时按下上下左右键的任一键,操纵器会以小单位移动.Shift + 操纵器正常会移动一小单位10um X-bzw, Y-RichtungCtril + 操纵器正常会移动一小单位250um X-bzw, Y-Richtung单位宽度可在cnc.ini档案中调整(INI-FileCNC.INI)附加打印数据对话框如要打印一个影像或一个显示区域, 以下附加数据可以一并印出.Inspector 检验者名称Production Line 生产线叙述Part.-No. 组件号码Job 工作描述Comments 附加测试描述字段Cancel 取消对话框,不需打印Print without data 打印影像不含附加信息Print 打印影像包含附加信息在页首打印信息手动比对调整对话框手动比对调整允许手动设定灰阶值的范围,其分布的范围0 .. 255.长条图(灰阶值的数量…灰阶值呈现在影像中的数值频率)的显示和灰阶值0…255的分布可由两个滚轴来设定. 如果鼠标光标移动滑过长条图, 灰阶值(Grey)和其总数(Count)会显示在右边.如果影像是实时的,长条图会定期的更新, 所以更换标的物或撷取参数(kV, uA, Gain,…) 会即刻反映出来.滚轴“low”决定较低的灰阶值而滚轴“high”其较高灰阶值0…255分布由. 较低灰阶值用红色代表,较高灰阶值在长条图中用绿线表示.如果滚轴的值, low=31, high=228, 此影像灰阶值会显示0..31为0而值228..255为255,中间值会成比例的分布在0到255的直线上.如果高数值低于低数值,表示其影像为反白的,所有灰阶值低于高的会显示为255而灰阶值高于低的会显示0.中间值会成比例的分布在0到255的直线上.Default 重置系统隐含值,low=0, high=255, 即没有提高反差比对.OK 接受提高反差比对并离开对话框Cancel 取消设定并离开对话框数字化设定此对话框能调整撷取卡结构的设定.(撷取器, 数字器)来连接X-Ray的影像系统.“Contrast” 和“Brightness” 控制摄影机模拟讯号的数字化.这是与反差对比的数字影像不同的.(例: Contrast Manual)Contrast应该要调整对于一个完全过度曝光的(饱和的)影像会造成所有的画素灰阶值为255, 用尽可能的低阶值来调整.Brightness应该要调整对于一个全黑(X-Ray 关闭)影像会造成所有的画素灰阶值为0,用尽可能的高明亮值来调整.可能的明亮值由0=黑到255=白可能的对比值由0=黑到511=白这些都只是表示值.实际上的设定要依赖未来的因素来设定VGA显示器的明亮度与反差对比的设定和个人对于影像的敏感度.Video Input能由连接摄影机的影像撷取卡来调整.根据标准是用input 1,S-VHS录像机是用Input S.如果系统是装配在OVHM modul, 您不能更动video input.Synclevel能经由调整影像撷取卡来要求video 讯号的质量, 例如: 由录像机或光盘片. Synclevel标准设定值为125mV.完成了所有的设定之后, 请按下Done的键来离开对话框.批注选择在窗口上方边缘的控制键,对话框会在右边跳出.在此您可以键入您所要的批注.批注会重迭在影像上,因此影像本身不会遭受破坏,除非您将批注贴入影像上.(请见列批注于图表上).由此批注可以依照不同形式更改,因此称为可变动批注.批注的型态Text (正文) 多面的正文内容,尺寸,形态和颜色,透明Line (线条) 任何尺寸,颜色和线条宽度Rectangle, Ellipse (矩形, 椭圆) 任何尺寸,形状,颜色和线条宽度Arrows (箭头)任何尺寸的箭头向着四个不同的45度角附加批注选择批注窗口的控制键可加入批注.此批注窗口会预设显示在影像左上角.提示: 加上正文批注并按住Ctrl-key, 现行的日期与时间即可加注.移动批注在批注上按住鼠标左键, 用鼠标来拖曳批注.重订批注的尺寸如果以个批注的尺寸可以重订,即控制点(小绿色矩形)会显示在矩形批注的周围.用鼠标左键按住控制点来改变尺寸.用角落的控制点来重订尺寸,其放缩比例是一定的.按住Ctrl-Key, 重订的尺寸是对称的.更改批注只要批注没有被贴在影像,即可依以下方式更改:按下鼠标右键或在批注上点两下即可看到一个内文窗口.依此状态表您可选择不同功能来更改批注.一般此功能可能适用在所有批注形式:To front将批注带到前景,如此就不能由其它批注所覆盖To back将批注带到背景,原来被覆盖的批注变得显而易见且可选择.Delete删除所选择的批注TextChange text显示一个对话框可以用来进入或改变批注矩形背景将会被调整到新的批注,如此新的批注就可显示出来Font调整字体,形态和尺寸.在增加字体尺寸之后,选择“change text” 来调整矩形背景Color改变批注的颜色Background color如果不是透明,可以改变批注背景的颜色.Transparent无背景下显示批注,只有影像.To front, To Back, Delete 请见以上.LinesColor改变线条颜色Line width改变线条宽度从1(细) to 9(粗)Line ends改变线条尾端Rectangle, Ellipse, ArrowsColor改变笔的颜色Line width改变线条宽度从1(细) to 9(粗)Ellipse (椭圆形)Rectangle (矩形)Arrow to right top(向右上方箭头)Arrow to let top(向左上方箭头)Arrow to left bottom(向左下方箭头)Arrow to right bottom(向右下方箭头)Set to default size(初始设定尺寸)选择一个形状,如果批注的尺寸改变,可重设这个尺寸在影像中贴上批注选择一个贴上的控制键,所有可更动的批注将可贴在影像中.执行此指令,影像的颜色会消失, 并由相对的灰色来代替.因此小心选择适当的颜色.你可得到一个黑色和白色的批注在影像上.正常来说,可更改的批注在贴到影像上之后将会被删除.如果需要,您可改变此动作一但黑白的批注已贴入图形后, 之后的修改就变得不可能且困难.(例: 更改日期…)删除所的批注选择waste basket(垃圾桶), 所有可更改的批注在您选择yes之后都将会被删除.储存批注储存一个含有批注的影像(批注不是贴在影像上的), 批注会依相同路径与文件名但不同的附加档名“.ano”储存.下载此影像文件附加文件名“.ano”的档案自然存在,可更改的批注同样也会被载入.Wire Sweep此为选备项目.如何操作Wire Sweep?Wire sweep 的功能是用来鉴定金线的弯度.由此可测出金线最弯曲点到两端直线垂直距离,对金线两端距离的百分比.如果选择此功能,鼠标光标会由x 标志的取代.请依序按下鼠标在线的两个末端, x 标志将会在线的两末端点上做上记号而且出现连接线.移动鼠标拉一条连接线到金线最弯的位置上,按鼠标左键.在此过程中, wire sweep的值会一直显示在右边wire sweep 状态表中. 如果wire sweep 的值是在标准下(请参考wire sweep 设定), 其显示值为绿色否则颜色会变成红色. 如此便可马上了解被测线的弯度是许可的或不许可…完成量测之后,一个对话框会显示Wire-Sweep 的值并分类出“Fail” 或“OK”.之后再量测下一条线.如果所有的线都已被量测完成请点选Wire Sweep 状态框或按下ESC键来离开Wire Sweep 的功能.Wire Sweep Setup此对话框可设定Wire Sweep 最大的有效值.如果线的Wire-Sweep值超过所设的标准值,此线会被判定为FAIL而Wire-Sweep-标准值的颜色会由绿色变成红色.量测Conducting a Measurement以上影像显示3点残留环宽量测. 量测结果是37um可由邻近的“measure”键中看见.此量测功能提供被测对象的距离和分离信息. 可选择三种量测方式: 二点,三点和CNC量测.二点量测, 其名称所示,决定两个使用者定义点的距离. 三点量测, 首先使用者需先在影像中设定两个点来定义出一条线. 第二条并行线会显示出来给使用者定位在量测点上, 此种量测方式在量测直径和残留环状时特别有用. 两种量测方式是分别将点或线的像素尺寸累积决定出真实的尺寸.因此,要紧的是像素尺寸正确的校正.对于没有CNC系统的使用者来说必须测定像素的尺寸.而对于有CNC系统的,参数会自动设定. CNC量测模式使用轴的位置会对应到像素尺寸,如此使用者被提供一个十字线来协助在屏幕中央定位需求点. 使用以上任何一种量测方式,仅要按下“Measure”键和选择所需求的量测方式即可.注意: 对于没有CNC 系统来说像素尺寸必须由使用者在指定的倍率来自行计算注意: 如果系统加装CNC, 最重要的是确定系统对影像接收器使用正确的Zoom倍率.两点量测点选右边所显示的符号. 当鼠标光标显示在影像中, 鼠标光标会转变成十字线. 点选两个点来量测, 这两个点会由一条线连接起来.在选择第二个点的位置时,两点间的距离会实时的出现在“Measure”键旁边的窗口中.三点量测点选右边所显示的符号. 当鼠标光标显示在影像中, 鼠标光标会转变成十字线. 现在设定两个点作为量测线.设定第一条线之后,再选择移到第二条并行线到量测点上.在定位好第二条线位置时, 两条线的间距会显示在“Measure”键旁边的窗口中.在用鼠标左键来确认第二条线的位置之后,一个对话框会显示出两条线径的距离. 量测单位可由“Setup Measurement”对话框中选择.在此后可以进行下一个量测或选择其它量测方式.如果不再需要量测,即可点选“Measure”键或按下“Escape” 键即可.CNC-量测如果系统软件有安装CNC, 就可选择此种量测功能. 要确认轴的绝对位置,回授机构是必需的.用鼠标左键来点选“Measure”. 然后点选右边所显示的符号. 一个实时影像和十字标志会显示出来,现在使用操纵杆移动到第一个位置并用鼠标左键确认,然后用同样方式移动到第二个位置并再用鼠标确认.用鼠标左键来确认线的位置后,一个对话框会显示出距离来.量测单位可由“Setup Measurement”对话框中选择.在设定第二条线位置时, 两条线的间距会显示在“Measure”键旁边的窗口中.如果不再需要量测,即可点选“Measure”键或按下“Escape” 键即可.量测设定以下设定需由管理层级的操作者来设定如果系统加装CNC, 此对话框是用来告诉软件系统,在量测方式中将要使用到的机器重要尺寸.如果系统未加装CNC, 以上信息除单位之外将不需理会.注意: 无论是否选择英吋单位,系统单位始终设定为mm.Unit定义量测单位.在量测标的物的厚度与像素尺寸时,也会被用来使用.Focus-II-Distance这个值是表示X-Ray管的焦点和影像接收器实际影像接收面的距离.而焦点和X-Ray 管出口的距离是依X-Ray 管而定. 130kV的值是18mm.影像接收器的铝盖板和接收面的距离是8mm.Focus-Object-Distance这个值是表示焦点和检验桌之间的距离.而焦点和X-Ray管出口的距离是依真空管而定. 130kV的值是18mm. 影像接收器的铝盖板和接收面的距离是8mm.Optional Factor 0/1 X/Y这个值是x/y 像素在影像接收器放说缩比例0到1尺寸的实际尺寸.这个值是依照摄影机像素尺寸(Pcam), 测量物(f), 影像接收器缩小比例(Mbv)来计算的:计算方式如下:Ox=Pxcam/f*/Mbv(y analog).经由不断测试,实际值将会被精确的订定.注意: 对于没有CNC 系统的,尤其重要的是光学对位的调校正确.。

Optical related technologyX-ray DiffractionAbstract:X-ray diffraction (XRD) technology is widely used in materials characterization. Identify chemistry constituents of the product, analyze the space group, lattice parameter and etc. In this article, I will introduce the principle of XRDincluding the diffractometer and simple application with the XRD pattern.Introduction1.X-ray is a kind of electromagnetic radiation generated by the inner electron transition. Its wavelength ranging from 0.01 to 10nm. X-ray photons with a short wavelength (below 0.2nm) have the energies above 5keV. Therefore comes the penetrating ability [1]. Moreover, X-ray interact with matter and project the matter’s information in X-ray diffraction pattern.As result, X-ray are widely used to image the inside of the object and is thought highly in the field medical radiography and material characterization.2.X-ray diffraction is an interaction between X-ray and matter.Consider the materialas lattice as show at figure 1. Consider thesituation as what we had learned-the Rayleigh scattering. Insome direction, we can detect the interference of two narrowX-ray beams, and constructive interference meet the opticalpath difference:2dsinθ=nλ (n=0,1,2…).Kwon as the Bragg’s law. d is the distance between two particles, θis the angle between incident X-ray and lattice plane,λis the wavelength of X-ray.We can learn from the equation that the constructive interonly happen in some particular angle since the wavelength and the distance(consider as the grating constant) are fixed for one material, and for the three dimensional world we have three constant marked as (h k l) to describe the space group. Nevertheless, is hard to directly detect the 3D pattern, instead, we introduce the spherical projection pattern and develop some mathematical translation to restructurethe 3D condition shows at figure 2. We canlearn the direction and distance from thespherical coordinate. Let’s go over theprocedure of XRD. First we shot a X-ray to thesample and get its diffraction pattern and readthe information from the 2D pattern[2].3.X-ray Diffractometer is high integration optical apparatus shows at figure 3. The X-ray source usuallychooses Cu as target material applying hot electrons whichaccelerated by high voltage and generate X-ray [3]. Sampleplatform and detector link together and rotate along thehorizontal spherical center axis. We can also have thesample platform and detector fixed and rotate the X-ray gun.We introduce the concept of resolution to evaluate thequality of the X-ray Diffractometer. A strong and stable X-raygenerator is necessary. The other condition is the minimumangle which the machine can stable rotate. The resolutionusually plays an important role in quantifying analysis suchas determine the doping ratio by measure the XRD peak offset.Simple analysis with XRD patternAssume we have got our XRD pattern file, we can import them into XRD analysis software “jade” shows at figure 4. As we already talk about that XRD can identify the space group of the crystal material. If we get the standard XRD pattern (verified by other method). We can compare our XRD pattern to the standard pattern (PDF) and see whether the diffraction peak match. In figure 4, I take the Lead sulfur (PbS) as an example. The mountain like curve is the XRD pattern we detect from the X-ray Diffractometer, x axis is 2θ/degree and the y axis represent the intensity; Vertical line is PDF, it marks the position and intensity of the standard diffraction pattern of PbS. So we can read the crystal parameter from the PDF information: space group Fm-3m (No.225), the size of the cell is 5.9362*5.95362*5.9362<90*90*90>we can know the shape of PbS cell is almost cube. But we still need SEM and TEM technology to precise determine, recall the XRD characterization procedure, we only learn the information from PDF. If you have a deeper look on the XRD pattern in figure 4, the XRD pattern peak unitary slightly offset to the left against thePDF’s. It suggest s the PbS sample has a smaller size than the PDF’s standard si ze. We can calculate the exact size by applying mathematics butI won’t go that further.Conclusion and expectationIn this article, we discussed the principle of X-ray, X-ray Diffraction, X-ray Diffractometer and X-ray diffraction pattern. It turns out the XRD technology is so convenient to image the inside of the matter. But we also see the shortcoming of this technique: 3D to 2D projection lose some information and fail to detect the absorption information. I suggest having more detector in different position to improve the accuracy and tunable X-ray source to determine the absorption of the material.Reference[1]https:///wiki/X-ray[2] <晶体学基础>秦善[3]X衍射以及其应用简介—陶琨（清华大学）。

a r X i v :a s t r o -p h /0309460v 2 16 S e p 2003Draft version February 2,2008Preprint typeset using L A T E X style emulateapj v.11/12/01THE NATURE OF E+A GALAXIES IN INTERMEDIATE REDSHIFT CLUSTERS 1,2Kim-Vy H.Tran 3Department of Astronomy &Astrophysics,University of California,Santa Cruz,CA 95064vy@phys.ethz.chMarijn FranxLeiden Observatory,P.O.Box 9513,2300RA Leiden,The Netherlandsfranx@strw.leidenuniv.nlGarth IllingworthUniversity of California Observatories/Lick Observatory,University of California,Santa Cruz,CA 95064gdi@Daniel D.KelsonObservatories of the Carnegie Institution of Washington,813Santa Barbara Street,Pasadena,CA,91101kelson@Pieter van DokkumDepartment of Astronomy,Yale University,New Haven,CT 06520-8101dokkum@ Draft version February 2,2008ABSTRACTCombining HST/WFPC2mosaics with extensive ground-based spectroscopy,we study the nature of E+A galaxies in three intermediate redshift clusters (z =0.33,0.58,&0.83).From a sample of ∼500confirmed cluster members,we isolate 46E+A candidates to determine the E+A fraction and study their physical properties.Spectral types are assigned using Balmer (H δ,H γ,H β)and [OII]λ3727equivalent widths.For all members,we have galaxy colors,luminosities,Hubble types,and quantitative structural parameters.We also include measured internal velocity dispersions for 120cluster members,and estimate velocity dispersions for the rest of the cluster sample using the Fundamental Plane.We find E+A’s comprise a non-negligible component (∼7−13%)of the cluster population at these redshifts,and their diverse nature indicates a heterogeneous parent population.While cluster E+A’s are predominantly disk-dominated systems,they span the range in Hubble type and bulge-to-total fraction to include even early-type members.Cluster E+A’s also cover a wide range in luminosity (L B ∼0.2−2.5L ∗B ),internal velocity dispersion (σ∼30−220km s −1),and half-light radius (r 1/2∼0.4−4.3h −1kpc).From their velocity dispersions and half-light radii,we infer that the descendants of E+A’s in our highest redshift cluster are massive early-type galaxies.In contrast to the wide range of luminosity and internal velocity dispersion spanned by E+A’s at higher redshift,only low mass E+A’s are found in nearby clusters,a.The observed decrease in the characteristic E+A mass is similar to the decrease in luminosity of rapidly star-forming field galaxies since z ∼1,i.e.galaxy “down-sizing.”In addition,we argue our statistics imply that 30%of the E-S0members have undergone an E+A phase;the true fraction could be 100%if the effects of E+A down-sizing,an increasing E+A fraction with redshift,and the conversion of spirals into early-types are also considered.Thus,the E+A phase may indeed be an important stage in the transformation of star-forming galaxies into early-type members.Subject headings:galaxies:clusters:general —galaxies:evolution —galaxies:fundamentalparameters —galaxies:structure —galaxies:high-redshift1.introductionPost-starburst galaxies (“E+A”;Dressler &Gunn 1983)in clusters may provide the crucial link in the morpholog-ical transformation of spiral galaxies into the elliptical/S0systems that dominate the cluster population at the cur-rent epoch (cf.Butcher &Oemler 1978).Characterized by strong Balmer absorption and little or no [OII]λ3727emission,E+A’s also are referred to as k+a/a+k (Franx1993;Poggianti et al.1999;Dressler et al.1999,hereafter D99)or H δ-strong galaxies (Couch &Sharples 1987;Couch et al.1994).Despite the short window of visibility of the post-starburst phase (<1.5Gyr;Couch &Sharples 1987;Barger et al.1996;Leonardi &Rose 1996),E+A members can contribute up to 20%of the total cluster population (D99),and have been found in virtually every spectro-scopic cluster survey from 0<z <0.8(Dressler &Gunn1Based on observations with the NASA/ESA Hubble Space Telescope,obtained at the Space Telescope Science Institute,which is operated by the Association of Universities for Research in Astronomy,Inc.,under NASA contract NAS 5-26555.2Based on observations obtained at the W.M.Keck Observatory,which is operated jointly by the California Institute of Technology and the University of California.3Current address:Institute for Astronomy,ETH H¨o nggerberg,CH-8093Z¨u rich,Switzerland121983;Couch&Sharples1987;Wirth et al.1994;Cald-well et al.1996;Balogh et al.1999;Dressler et al.1999; Poggianti et al.1999;Tran2002).However,the fraction of E+A galaxies in intermediate redshift clusters(0.2<z<0.8)is still debated.The MORPHS survey estimated a high E+A fraction of∼20% (D99)while the CNOC1survey found a much lower frac-tion of∼2%(Balogh et al.1999,hereafter B99).These results conflict as to whether the majority of cluster mem-bers undergo an E+A phase,or if only a small fraction do. If the real E+A fraction in clusters is high,it can be an important constraint on how members have evolved.For example,the transformation of late-type spirals via a post-starburst episode may explain the high number of passive S0galaxies seen in nearby clusters(Dressler&Gunn1983; Couch&Sharples1987).Once a robust E+A sample in intermediate redshift clus-ters is established,more fundamental issues such as the nature of the parent population can be addressed.Past studiesfind that while E+A’s are heterogeneous in mor-phology,they tend to have disks(Wirth et al.1994;Couch et al.1994;Caldwell&Rose1997;Couch et al.1998,D99). Rotation velocities determined for a small subset of clus-ter E+A’s at z 0.3also show most of them are rota-tionally supported systems(Franx1993;Caldwell et al. 1996;Kelson et al.2000b,hereafter K00b).However,it has been suggested that manyfield E+A’s and some clus-ter E+A’s are the result of processes that would easily disrupt a disk,e.g.merging and/or strong galaxy-galaxy interactions(Belloni et al.1995;Liu&Kennicutt1995; Zabludoffet al.1996).Whether cluster E+A’s include even the most massive galaxies,as suggested by Wirth et al.(1994),remains an open question.A more challenging but extremely interesting question is what the descendants of cluster E+A’s are.To answer this question,diagnostics that are not likely to evolve strongly with redshift are needed,e.g.internal velocity dispersions (σ)and half-light radii(r1/2).However,published work on E+A’s that utilizes bothσand r1/2have been limited to clusters at z 0.3(Franx1993;Caldwell et al.1996; Kelson et al.1997,2000c,hereafter K00c).By applying a similar analysis to E+A’s at higher redshifts(z 0.3),we can determine if even the most massive galaxies in nearby clusters had an E+A phase in their past.Only with a com-bination of deep spectroscopy and high resolution imaging can both these parameters be measured with confidence in intermediate redshift clusters.In thefield,Cowie et al.(1996)find the maximum lumi-nosity of galaxies undergoing rapid star formation has de-creased smoothly since z∼1.This evolution in the char-acteristic mass of the star-forming population,referred to as“down-sizing,”reflects how mass assembled,i.e.more massive galaxies formed at higher redshift.Recent work in clusters suggest down-sizing also can play an important role in rich environments(Poggianti et al.2001;Kodama &Bower2001;Poggianti2003).If this is the case,the lu-minosity and internal velocity dispersion distributions of cluster E+A’s should evolve as a function of redshift.By determining if the characteristic mass of cluster galaxies undergoing the E+A phase evolves,we can provided ad-ditional compelling evidence for down-sizing in clusters. The high fraction of E+A galaxies in clusters relative to thefield suggests that environment plays a significant role in producing E+A’s(D99).For example,the E+A phase may be triggered by an external source such as the strong galaxy-galaxy interactions suggested by studies of field E+A’s(Liu&Kennicutt1995;Zabludoffet al.1996). In addition to direct mergers(van Dokkum et al.1999), the cluster environment provides numerous other exter-nal forces that could trigger an E+A phase,e.g.ram-pressure stripping(Gunn&Gott1972),gas compression (Dressler&Gunn1983),perturbation by the tidalfield (Byrd&Valtonen1990),and galaxy harassment(Moore et al.1996).The more disruptive interactions would cre-ate morphological signatures that should be visible during the E+A lifetime.Examining the number of E+A’s that show morphological signs of recent mergers or interactions can help isolate which interactions trigger the E+A phase in clusters.Also,by comparing their spatial distribution to cluster substructure,we can test if the global cluster en-vironment is effective at producing E+A’s,or if they are better correlated with local over-densities.To address these issues,we mustfirst overcome the in-herent difficulty of isolating E+A galaxies.A statisti-cally representative sample of members(>100)in each cluster is needed so that the relatively few E+A galax-ies can be identified.Membership confirmation and E+A selection can only be accomplished reliably by obtaining spectra.In addition,high resolution imaging is needed at these redshifts(z>0.3)to determine physical prop-erties such as structural parameters and morphological type.Only by pairing wide-field HST/WFPC2imaging with deep ground-based spectroscopy can we adequately study the E+A galaxies in intermediate redshift clusters. From extensive spectroscopic surveys of CL1358(z= 0.33;Fisher et al.1998),MS2053(z=0.58;Tran2002), and MS1054(z=0.83;Tran et al.1999;van Dokkum et al. 2000;Tran2002),we select46E+A candidates from∼500 confirmed cluster ing HST/WFPC2mosaics taken of each cluster(all to R BCG∼1h−1Mpc),we mea-sure the colors,magnitudes,half-light radii,bulge-to-total fractions,degree of galaxy asymmetry,and morphologi-cal type of the cluster members.With LRIS(Oke et al. 1995)on Keck,we also have measured internal velocity dispersions for120cluster members(Kelson et al.1997; van Dokkum et al.1998a;Kelson et al.2001,2003,K00b). With these measured dispersions,accurate colors,and the Fundamental Plane(Faber et al.1987;Djorgovski&Davis 1987),we estimate velocity dispersions for the remainder of the ing the E+A’s that satisfy our strict selection criteria,we determine the E+A fraction in inter-mediate redshift clusters,identify characteristics of their parent population,address what the descendants of these galaxies can be,and discuss the likely down-sizing of this population.A brief description of the HST/WFPC2imaging and ground-based spectroscopy is provided in§2.We describe our E+A selection criteria and address the discrepancy in the cluster E+A fraction found by different surveys in§3. After identifying the cluster E+A population,we examine the nature of these systems in§4.We discuss their prop-erties and evolution in§5,and present our conclusions in §6.Unless otherwise noted,we useΩM=0.3,ΩΛ=0.7, and H0=100h km s−1Mpc−1in this paper.32.summary of observations and dataWe select E+A galaxies from a large program studying three X-ray luminous clusters at z=0.33,0.58,&0.83 (Table1,references therein).Our dataset combines HST/WFPC2mosaics(each to R BCG∼1h−1Mpc)of these clusters with extensive ground-based spectroscopy. From the spectroscopy,we determine spectral types for all members(Fisher et al.1998;van Dokkum et al.2000;Tran 2002)as well as measure internal kinematics for a subset (Kelson et al.1997;van Dokkum et al.1998a;Kelson et al. 2001,2003,K00b).From over1200redshifts obtained in the threefields,we isolate∼500cluster members.Here we describe briefly the spectra and photometry used in this paper.2.1.PhotometryThe three clusters were imaged by HST/WFPC2in the F606W and F814Wfilters.The image reduction and pho-tometry are detailed for CL1358,MS2053,and MS1054in van Dokkum et al.(1998b),Hoekstra et al.(2000),and van Dokkum et al.(2000)respectively.Following the method outlined in van Dokkum&Franx(1996),we transform from the WFPC2filter system to redshifted Johnson mag-nitudes using:B z=F814W+a(F606W−F814W)+bV z=F814W+c(F606W−F814W)+d(1) where the constants{a,b,c,d}for an E/S0galaxy are {1.021,0.524,0.204,0.652}z=0.33{0.354,0.923,−0.173,0.959}z=0.58{−0.077,1.219,−0.524,1.229}z=0.83(2)These apparent magnitudes correspond to integrating the galaxy’s spectral energy distribution through the red-shifted Johnsonfilter curves.Apparent magnitude then is converted to an absolute magnitude adjusted for pas-sive evolution(M Be)using the distance moduli in Table1; here we account for simple fading,as determined from the Fundamental Plane(∆log(M/L)∝−0.40z;van Dokkum et al.1998a).In our analysis,we include Hubble types from Fabricant et al.(2000,2003)who visually typed all members with m814<22.In the nomenclature adopted by these authors, the morphological types of{E,E/S0,S0,S0/Sa,Sa,Sb,Sc, Sd}were assigned values of{−5,−4,−2,0,1,3,5,7};inter-mediate values of{−3,−1,2}were also used and mergers assigned a value of99.In the following,we consider E-S0galaxies as having−5≤T≤−1,S0/a-Sa galaxies 0≤T≤1,and spirals2≤T≤15.2.2.SpectroscopyIn CL1358,spectra of230members were obtained at the WHT and MMT.Details of the target selection and spec-tral reduction are in Fisher et al.(1998,hereafter F98). Additional high resolution spectra were taken with LRIS (Oke et al.1995)on the Keck Telescope to obtain internal velocity dispersions of55members(K00b).Keck/LRIS was also used to obtain redshifts and internal velocity dis-persions for both MS2053and MS1054.In MS2053and MS1054,150and130members were confirmed respec-tively(Tran et al.1999;van Dokkum et al.2000;Tran 2002),and internal velocity dispersions for29and26mem-bers measured(Kelson et al.1997;van Dokkum et al. 1998a;Kelson et al.2003).Detailed explanations of the spectral reductions including the wavelength calibration, sky subtraction,and removal of telluric absorption can be found in the noted references.The wavelength coverage for virtually all members in the three clusters includes[OII]λ3727,the4000˚A break,and Balmer lines(Hδ,Hγ,&Hβ).The inclusion of these spec-tral features are of paramount importance as E+A galaxies are defined spectroscopically as having strong Balmer ab-sorption and no[OII]λ3727emission(Dressler&Gunn 1983;Zabludoffet al.1996;Balogh et al.1999;Poggianti et al.1999,D99).The bandpasses used to determine the equivalent widths of these features are listed in Table2.3.defining e+a galaxies3.1.E+A Selection CriteriaWe select E+A galaxies from the three clusters in our sample(Table1)as having(Hδ+Hγ)/2≥4˚A and no[OII] emission(>−5˚A).Although using all three Balmer lines to select E+A’s is the most robust approach(Newberry et al.1990),Hβis severely compromised by sky lines in MS1054and so we do not include Hβin our selection cri-teria(see Table3).Due to the wavelength coverage,we cannot determine the equivalent width of[OII]λ3727for ∼7%and∼4%of members in MS2053and MS1054re-spectively;for these galaxies,only Hδand Hγare used to determine their E+A status.This adds four E+A’s (H3549,H2345,H1746,H408)to the MS2053sample.The 46cluster members that satisfy these criteria are shown in Fig.1and their physical characteristics are listed in Ta-ble4.Because the spectral quality varies for the three clus-ters,we define a magnitude limit(M Be≤−19.1+5log h, m814∼21.9)set by MS1054,the highest redshift cluster in our sample.For uniformity,we consider only members brighter than this cut when comparing the E+A popu-lation between the clusters.In selecting E+A’s,we also apply a signal to noise cut(S/N≥20)on the Hδand Hγfluxes.Only14of the46E+A candidates satisfy these strict selection criteria;these spectra are shown in Fig.2. An important point is that the spectroscopic criteria used to identify E+A galaxies vary depending on author, as can be seen in Table3.The use of different thresholds and lines,e.g.only Hδversus a combination of Balmer lines,can produce different E+A fractions even within the same sample.The most robust approach is to use all three Balmer lines(cf.Newberry et al.1990)and[OII]λ3727 but,as noted earlier,this is not possible for our entire sample.Thus we restrict ourselves to using[OII]λ3727in combination with Hδand Hγ.The only exceptions in our sample are2053–3549and2053–2345.Although we do not have OII EQW’s for these two E+A candidates,both have strong Hβabsorption such that(Hδ+Hγ+Hβ)/3>4˚A, and all three Balmer lines have S/N>20.It is very unlikely that these two E+A candidates also have strong OII emission because in the CL1358and MS2053samples, >90%of members with(Hδ+Hγ+Hβ)/3>4˚A show no significant[OII]λ3727emission.We confirm1054–6567’s lack of OII emission from a lower S/N spectrum.4Figure3shows(B−V)z versus(Hδ+Hγ)/2for the cluster sample;morphological types are included(Fabri-cant et al.2000,2003).If we consider only the galaxies brighter than our imposed magnitude limit,the average color of E+A galaxies tends to be bluer than that of the early-type population in each cluster.As simple errors in the spectroscopy could not produce such a uniformly blue sample in all three clusters,the E+A fraction must be real and robust.3.2.The E+A Fraction in Intermediate Redshift Clusters For members brighter than our magnitude cut(M Be=−19.1+5log h),we obtain E+A fractions of7±4%,10±6%, and13±5%at z=0.33,0.58,&0.83respectively(Ta-ble5);note we apply our strict E+A selection criteria here.If we include all confirmed members in each cluster and include all E+A candidates,the fractions are9±2%, 7±2%,and16±3%.Errors are determined by assuming a Poisson distribution for the E+A galaxies. Considering the low E+A fraction in Coma4( 3%; Caldwell et al.1993),these results show the E+A fraction evolves strongly with redshift.They also suggest the E+A fraction continues to increase at z>0.5.However,larger samples at z>0.3are needed to determine if this increase is real or if the trendflattens at z>0.3.parison to CNOC1As noted earlier,there is disagreement in the literature as to whether the E+A fraction in intermediate redshift clusters(z 0.3)is significantly higher than that of the field.Published cluster surveys(0.3<z<0.6)estimate cluster E+A fractions of∼10−20%(Belloni et al.1995; Couch et al.1998,D99).In contrast,B99argue that the average cluster E+A fraction is negligible(1.5±0.8%)at z=0.18−0.55,and comparable to thefield.Is it possible to reconcile these two remarkably different claims? While both D99and B99spectroscopically select their E+A samples using Hδand[OII]λ3727,B99apply a cor-rection that decreases their cluster E+A fraction.Rec-ognizing that quiescent galaxies with low Balmer indices dominate the cluster populations,B99argue measurement errors will automatically produce outliers that are classi-fied as E+A’s.To remove this artificial inflation,they cor-rect the fraction of“raw”cluster E+A’s from4.4±0.7% to1.5±0.8%.Furthermore,B99note that in CL1358the E+A fraction determined by F98(4.7±1.9%)is remark-ably similar to the uncorrected E+A fraction from their combined cluster sample(4.4±0.7%).They suggest F98 overestimated the true fraction by not correcting for in-clusion of spurious outliers.We test B99’s theory by focusing on CL1358,a cluster included in B99’s analysis and spectroscopically surveyed by both CNOC1(Yee et al.1996)and F98.If a signif-icant number of E+A’s in this cluster are actually pas-sive galaxies,many of the E+A’s should be as red as the early-types.Figure4shows the distribution of color ver-sus Balmer strength for cluster members taken from F98 and B99;note B99use only Hδas their Balmer criterion while F98use the average of Hδ,Hγ,and Hβ(Table3).While E+A’s selected by B99have a large spread in color and can be as red as the passive population,virtually all of the E+A’s found by F98are bluer than the passive early-types.Simple errors in the spectral indices could not pro-duce such a uniformly blue sample,and so these cannot be spectroscopically passive galaxies mistaken as E+A’s. In comparing results from F98and B99for∼100com-mon members,wefind significant offsets in Hδand[OII] between the two surveys;this may explain the discrep-ancy in the E+A fractions.On average,Hδequivalent width values are smaller(∼0.7˚A)and[OII]emission larger(∼2˚A)in B99as compared to F98.Given the equivalent widths used to determine the E+A phase(see Table3),these offsets can seriously affect thefinal E+A selection.For example,if the indices of F98are trans-formed to the system of B99and B99’s selection criteria used,the fraction of E+A’s in CL1358as measured with F98’s spectra decreases from4.7%to3.4%.B99may have been motivated to correct their cluster E+A fractions if large errors were associated with the CNOC1Hδmeasurements.To determine if the CNOC1 errors are significantly larger than those of F98for CL1358, we independently estimate the equivalent width errors of both samples by comparing them to the high signal-to-noise spectra of ing27members common to the three studies,we estimate the true formal errors for Hδin both B99and F98are comparable(∼1.5˚A).This value is much smaller than the Hδformal error of2.8˚A associated with the F98sample and subsequently used by B99.Thus in CL1358,B99overestimated the correction factor due to measurement errors.To summarize,the discrepancy between B99and F98is due to1)the systematic offsets in the spectral indices,i.e. B99’s HδEQW values are smaller and[OII]λ3727larger than F98’s and2)B99’s overestimate of the correction due to measurement errors.As we have shown,the offsets in the spectral indices used to define the CNOC1CL1358 E+A sample reduces the observed number of E+A’s.This combined with B99’s correction factor produces a very low cluster E+A fraction.The E+A fraction(∼5%)in CL1358measured by F98using their stricter selection cri-teria is valid.4.nature of cluster e+a galaxies Having established that a significant fraction of E+A galaxies exists in clusters up to z∼0.8,we now exam-ine their physical properties to determine what type of galaxies they are.In the following,values for quantitative structural parameters,e.g.bulge-to-total fraction(B/T), galaxy asymmetry(R A),total residual(R T),half-light ra-dius(r1/2),and bulge/disk scale length(r e,r d),were mea-sured byfitting two-dimensional de Vaucouleurs bulge plus exponential disk surface brightness models to the galaxies (Tran2002;Tran et al.2003).The galaxy residuals R A and R T are measured by by taking the difference between the HST images and best-fit r1/4bulge+exponential disk model;how the residuals are measured is explained more fully in Tran et al.(2003).Unless otherwise noted,we only consider cluster members above our magnitude cut4Note that Caldwell et al.(1993)selected E+A’s using different selection criteria from a sample of predominantly early-type members.How-ever,recent results from the WINGS survey(Poggianti et al.2001;Poggianti2003)confirm the low E+A fraction in Coma for members brighter than M v=−18.5.5of M Be=−19.1+5log h,and only the E+A’s that satisfy our strict selection criteria.4.1.MorphologyIn this sample,the cluster E+A’s span the range in Hub-ble type(Fig.5)and bulge-to-total fraction(Fig.6)to in-clude both spirals and E/S0’s.However,the majority of E+A’s have measurable disks,consistent with results from previous cluster E+A studies(Wirth et al.1994;Couch et al.1994;Caldwell&Rose1997;Couch et al.1998;Cald-well et al.1999,D99).Their average bulge-to-total frac-tion of∼0.4reflects their tendency to be disk-dominated systems.This diverse range in B/T and Hubble type is similar to the heterogeneous morphologies found in studies of lower redshift cluster E+A’s(z 0.3;Couch et al.1998;Cald-well et al.1999).It also confirms suggestions in earlier studies that E+A’s must have a wide variety of progen-itors(Wirth et al.1994;Zabludoffet al.1996,D99).In addition,the earliest-type E+A’s are in our most distant cluster.It may be that more massive cluster members,i.e. early-types,had their E+A phase at higher redshift.4.2.Interactions&MergersThe morphological variety of E+A galaxies suggests that they are triggered by an external source,e.g.via the strong galaxy-galaxy interactions or the tidal forces proposed in earlier studies(Liu&Kennicutt1995;Bel-loni et al.1995;Zabludoffet al.1996;Caldwell et al. 1999).The cluster environment provides a plethora of possible disruptive mechanisms,e.g.ram-pressure strip-ping(Gunn&Gott1972),gas compression(Dressler& Gunn1983),perturbation by the cluster tidalfield(Byrd &Valtonen1990),and galaxy harassment(Moore et al. 1996).Also,the high merger fraction in MS1054indicates galaxy-galaxy merging is possible between members with low relative velocities(van Dokkum et al.1999).As these cluster E+A’s are predominantly disk-dominated systems, the more disruptive interactions would create morphologi-cal signatures(e.g.Barnes&Hernquist1992;Moore et al. 1998)that are visible during the E+A lifetime(∼1.5Gyr; Couch&Sharples1987;Barger et al.1996;Leonardi& Rose1996).By examining the number of E+A’s that are considered mergers and/or that have high galaxy residu-als,we attempt to isolate which interactions,if any,are associated with the E+A phase.To identify mergers,we use classifications from van Dokkum et al.(1999),Fabricant et al.(2000),and Fab-ricant et al.(2003).We consider high residual galaxies as those having a high degree of asymmetry(R A≥0.5; Schade et al.1995)and/or total residual(R T≥0.1;Tran et al.2001).Despite MS1054’s high merger fraction(∼17%;van Dokkum et al.1999),only two of the mergers are consid-ered E+A’s;no other E+A’s in our sample are associated with mergers.Wirth et al.(1994)and D99also observe a low incidence of mergers associated with cluster E+A’s. Wefind only about half of the cluster E+A’s have high galaxy residuals(Fig.6).For comparison,the fraction of high residual E+A’s is larger than that of early-types(E-S0;<15%)and even early-type spirals(S0/a-Sa;∼30%) but less than that of cluster spirals(∼80%).The num-ber of cluster E+A’s with prominent disks combined with only half the sample having high galaxy residuals suggests mergers are not the primary trigger of the E+A phase in cluster galaxies.4.3.Color-Magnitude DiagramEven though galaxies can be brightened significantly during the E+A phase(up to∼1.5mag;Barger et al. 1996),E+A’s in nearby clusters tend to be faint systems (L 0.4L∗;Caldwell et al.1999).However,past studies of intermediate redshift clustersfind E+A’s with L>L∗(Wirth et al.1994,D99).Here we determine if the clus-ter E+A’s in this sample are as luminous as the brightest cluster members,and whether they also cover a wide lu-minosity range.In Fig.7,we show the color-magnitude(CM)distribu-tion of all cluster members and E+A candidates;all clus-ter members,including the E+A candidates,have been corrected for simple passive evolution(§2.1).In all three clusters,wefind bright E+A’s(M Be −19.1+5log h); half of the14robust E+A’s are brighter than M∗Be (−19.5+5log h at z=0.83;Hoekstra et al.2000).Even more striking are the very luminous E+A’s at z=0.83: these E+A’s are up to a magnitude brighter than their lower redshift counterparts and cover a larger magnitude range.Note the cluster E+A’s tend to be bluer than the red sequence.The E+A luminosity range,particularly at z=0.83,only reinforces the conclusion that they have a heterogeneous parent population.The fact that the brightest E+A’s in this sample are in our most distant cluster is additional evidence for down-sizing of the clus-ter E+A population.4.4.Brightening During the E+A Phase Simple models show that during the post-starburst phase,galaxies can be brightened up to1.5magnitudes in the optical(Newberry et al.1990;Barger et al.1996). For comparison,we place here observational constraints on ∆M Be using the internal velocity dispersions(σ)acquired for120members and the Fundamental Plane(Djorgov-ski&Davis1987;Faber et al.1987,see Appendix).As demonstrated in e.g.Faber et al.(1987),residuals from the FP can be expressed as residuals in the M/L ratio. Wefind the log(M/L)residuals of the nine cluster E+A’s with measuredσrange from∼−0.5to∼0.3(Fig.8). Assuming E+A’s fade until∆log(M/L)=0,we estimate cluster E+A’s are brightened by as much as∆M Be∼1.25 mag,with a median of0.25mag.Assuming the E+A’s at z=0.83redden and fade by ∼0.25mag by z=0.33,the only galaxies in CL1358in this luminosity and color range are E-S0’s and S0/a-Sa’s (Fig.7).This suggests that some of the brightest early-type galaxies in nearby clusters had an E+A phase in their past.4.5.Internal Velocity DispersionsHaving demonstrated that E+A’s at higher redshift can be as luminous as the brightest cluster members(§4.3),we now determine if these brighter E+A’s are also massive galaxies,or whether they are simply low luminosity/mass members that are temporarily brightened.By determin-ing the E+A mass distribution,we can characterize what。

X-ray繞射基礎原理介紹演講者：王如春教授紀錄：黃信明什麼叫X-ray crystallography？即利用X光來定結構，這個結構包含了礦物結構、化學結構及蛋白質結構。

現在我們都知道X-ray是一個電磁波，為何要用x-ray最主要是和其波長有關，他的波長大約在一個Å左右。

Crystallography這個字是從crystal來的，這字的意思是一個整齊排列的物質，我們今天所用的工具是x光，對象是晶體，所謂的晶體是分子在固相中以整齊的方式排列，在理論上，固態和氣態是比較容易研究的，因為氣態的分子比較小，比較容易看單一分子的行為；固相中，在crystal，及分子是重複的整齊排列時，知道重複的單位，整個晶體的行為就可以瞭解；一般而言，液相的最難研究。

所以crystallography 就是利用X光來研究分子堆積在crystal狀態下的行為。

X光發現是在十九世紀末倫琴所發現，所以crystallography是一個跨領域的，當物理學家發展理論後，礦物學家就拿來作分析，因為礦物的分子最簡單所以最先使用，在二十世紀初就定出NaCl的結構。

隨著科技的發展就發現有更多分子的結構可以被分析出來，此時化學就被牽扯進去了，真正應用到biochemistry是在1980年代後期，此時我們才有足夠的設備和x光來做此研究。

在這一百年左右的歷史，可以看到理論的研究，到小分子的礦物最後研究到大分子的蛋白質分子。

現在開始介紹基本的繞射概念。

X光為何可以作結構分析，就取決於一個物理現象叫diffraction，當光和物質作用後的形式，如將一個石頭丟到水塘會產生一個波動，當波動撞到牆壁會反射回來，這時候產生干射的行為，就是說當波長非常長時，碰到一個平壁就會被反彈回來，如果波長很短，這時候大部分的波長會直接穿過，簡單的以物理概念來說，用一個狹縫寬度叫d，若波長比d要長，則這個波長就穿不過去會被彈回來，若比d小則會穿過去，若波長和d差不多大小會產生diffraction現象，要產生diffraction的現象得波長和光柵的距離得差不多。

rixs原理RIXS（Resonant Inelastic X-ray Scattering）是一种基于共振效应的非弹性X射线散射技术，广泛应用于研究固体材料的电子结构和元激发态。

它结合了X射线吸收光谱（XAS）和非弹性散射（IXS）的优点，可以提供关于元激发和晶格动力学等方面的详细信息。

下面是关于RIXS原理的相关参考内容。

1. 原子核共振在RIXS实验中，X射线的能量选择性地与特定元激发的共振能级相匹配。

这种共振可以是由于光子与原子核间的相互作用导致的，也可以是与电子间或是同位素间的相互作用导致的。

通过调整X射线能量以使其与共振能级相匹配，可以实现对元激发的选择激发和探测。

2. 能量损失谱RIXS实验中的主要观测量是能量损失谱（EL），它描述了样品吸收X射线能量后释放出的能量差。

EL谱可以提供有关元激发和电子结构的信息，包括能带结构、价电子态激发、自旋-轨道耦合、磁性和原子核动力学等。

3. 荧光增强RIXS实验通常使用荧光增强技术来提高信号强度。

荧光增强是一种通过在样品表面附近放置一个荧光标记来增强信号的方法。

荧光标记能够吸收入射X射线并发射出探测X射线，增强了信号的强度，并且减少了散射噪音。

4. 探测器和数据分析RIXS实验需要高分辨率的探测器来记录发射能量。

常用的探测器包括多道分析器和琼斯乃斯卡克探测器（Johannsson-Kristofel detector）。

实验数据需要经过精确的模型拟合和数据分析才能从中提取出元激发和结构信息。

5. 应用领域RIXS技术在许多研究领域中得到了广泛应用。

在固体材料中，它可以用来研究电荷传输、磁性、超导性和电子-声子相互作用等现象。

在凝聚态物理、材料科学和催化剂研究中，RIXS可以提供有关材料电子结构和元激发态的详细信息。

此外，RIXS还可以应用于生物、化学和环境科学等领域，以研究分子和化学反应。

总结：RIXS是一种关键的实验技术，可以提供关于固体材料的电子结构和元激发态的详细信息。

XGT-9000XRF Analytical MicroscopeScreen, Check, Map and MeasureThe combination of elemental images and transmission images allows one to detect hidden defects.Large working distance and coaxial vertical optics provide a clear transmission image without the shadow effect in undulating electronic boards.with elemental image only)and identifiedLine profile of blue part What is the XGT-9000?Screen, check, map and measureThe XGT-9000 is an X-ray Fluorescence Analytical Microscope, which provides non-destructive elemental analysis of materials.Incident X-ray beam is guided towards thesample placed on the mapping stage.X-ray fluorescence spectrum and transmission X-ray intensity are recorded at each point.Information available: Qualitative & quantitative elemental analysis/Mapping/Hyperspectral imaging.123Optical image Elemental imagesTransmission image Elemental imagesTransmission imageTransmission imageFull spectrum at each pixelfil f blXYThe XGT-9000 can detect anddetermine the composition of foreign particles, and therefore track the source of contamination.X-ray Fluorescence photons can be partially absorbed by theencapsulated material and will not show in the spectrum. The X-ray transmission image provides a complete picture.XGT-9000 with a wide range of applicationsOptical imageTiCrFeX-ray backscatter imageX-ray transmission imageAu thicknessOptical imageMapping areaLayered imageAu patternThe combination of microbeam and thickness measurement capability makes the XGT-9000 a useful tool for the QC of semiconductors,which feature thin and narrow patterns. Thickness sensitivity depends on elements traced, but can be at the Angstrom level.Biological samples contain water or gas, and will be heavily modified or damaged if measured in a vacuum. The unique partial vacuum mode of the XGT-9000 keeps the sample in ambient conditions while the detection is in a vacuum for optimum light elements measurement.Archeological artifacts are valuable materials and can only be analyzed by non-destructive methods.Dragonfly eye: XGT-9000 measurement has helped to ascertain the Dragonfly eye found in China actually originated Egypt/Middle East during the 2nd century B.C.Sample: Foreign matter in thecapsuleSample: Fly5 c mAlCaCu ComImage processing for mappingStandard GUI RoHS mode GUI Raw imageFloating viewQueue functionMultiple measurements including mapping /multi pointsResult list viewOptical imageParticle detectionFe image Particle detectionEdited GUIProcessed imageThe user interface offers a flexible way to measure multiple samples or areas in unattended mode (queue function),display the analytical results, present the data, and edit reports. Advanced treatments include image processing, particle finder, colocalization measurement and multivariate analysis (refer to "Combination of XRF and Raman Spectroscopies").XGT-9000 Software SuiteThe particle finding function is available from all the 3 images in the XGT-9000 (Optical, Fluorescence X-ray and Transmission). The particle finding function automatically detects particles and marks their position for multi-point measurement, classification and analysis.Coordinates of detected particles are automatically stored and transferred to the multi-point analysis modeViewbaTeh t s ak c a t S dn a p x ELabSpec linkCombination of XRF9 samples For 2”/4” wafersLow backgroundXGT-9000SLThe XGT-9000SL provides a non-destructive analysis of your most valuable pieces, which may be large or fragile.MESA-50 seriesElemental analysis and RoHS characterizationSLFA seriesThe reference instrument for sulfur-in-oil analysisIn/On-line solutionsReal time analysis forthickness and compositionDo more with your HORIBA XRFHORIBA XRF family* The sample chamber of the XGT-9000SL complies with the radiation safety requirement. The sample is measured in ambient conditions, while the detector operates at ambient or vacuum modes.XRF and Raman spectroscopies are complementary techniques.XRF provides information about elemental composition of the material, whereas Raman spectroscopy offers molecular information.Co-localized measurements between the XGT-9000 and HORIBA Raman spectrometers provide more information about the sample.Transfer of the XGT-9000 data to the advanced LabSpec Suite software using LabSpec link.Various sample holders areprovided to fit different shapes and types of samples.Fast and easy change between holders with HORIBA's modularstage design.Customization examplesTransfer vessel:Measurement of samples isolatedfrom airDimensionsXGT-9000SLXGT-9000(Unit: mm)(134)(476)(38)(9)(50)(1500)(2640)(1090)(1837)(1616)(74)(12)(16)(159)(769)74(2400)1800)003()003()A E R A E C N A N E T N A M()A E R A E C N A N E T N A M((3)(MANTENANCEAREA)DOOR OPENED914.5Bulletin:HRE-3764Ba Printed in Japan 2002SK62 The specifications, appearance or other aspects of products in this catalog are subject to change without notice.Please contact us with enquiries concerning further details on the products in this catalog.The color of the actual products may differ from the color pictured in this catalog due to printing limitations.It is strictly forbidden to copy the content of this catalog in part or in full.The screen displays shown on products in this catalog have been inserted into the photographs through compositing.All brand names, product names and service names in this catalog are trademarks or registered trademarks of their respective companies.3 Changi Business Park Vista #01-01, Akzonobel House,Singapore 486051Phone: 65 (6) 745-8300 Fax: 65 (6) 745-8155Unit D, 1F, Building A, Synnex International Park, 1068 WestTianshan Road, 200335, Shanghai, ChinaPhone: 86 (21) 6289-6060 Fax: 86 (21) 6289-5553Beijing Branch12F, Metropolis Tower, No.2, Haidian Dong 3 Street, Beijing,100080, ChinaPhone: 86 (10) 8567-9966 Fax: 86 (10) 8567-9066Guangzhou BranchRoom 1611 / 1612, Goldlion Digital Network Center,138 Tiyu Road East, Guangzhou, 510620, ChinaPhone: 86 (20) 3878-1883 Fax: 86 (20) 3878-1810Head Office2 Miyanohigashi-cho, Kisshoin, Minami-ku, Kyoto, 601-8510, JapanPhone: 81 (75) 313-8121 Fax: 81 (75) 321-5725HORIBA, Ltd.HORIBA Instruments (Singapore) Pte Ltd.HORIBA (China) Trading Co., Ltd.JapanSingaporeChina HORIBA India Private LimitedHORIBA (Thailand) LimitedIndiaTaiwanThailandPT HORIBA Indonesia Indonesia9755 Research Drive, Irvine, CA 92618, U.S.A.Phone: 1 (949) 250-4811 Fax: 1 (949) 250-0924HORIBA New Jersey Optical Spectroscopy Center20 Knightsbridge Rd, Piscataway, NJ 08854, U.S.A.Phone: 1 (732) 494-8660 Fax: 1 (732) 549-5125Via Luca Gaurico 209-00143, ROMAPhone: 39 (6) 51-59-22-1 Fax: 39 (6) 51-96-43-34Neuhofstrasse 9, D_64625, BensheimPhone: 49 (0) 62-51-84-750 Fax: 49 (0) 62-51-84-752016-18, rue du Canal, 91165, Longjumeau Cedex, FrancePhone: 33 (1) 69-74-72-00 Fax: 33 (1) 69-09-07-21HORIBA FRANCE SASGermanyFranceHORIBA Jobin Yvon GmbHItalyHORIBA ITALIA SrlHORIBA Instruments Incorporated USA246, Okhla Industrial Estate, Phase 3 New Delhi-110020, IndiaPhone: 91 (11) 4646-5000 Fax: 91 (11) 4646-5020Bangalore OfficeNo.55, 12th Main, Behind BDA Complex, 6th sector, HSR Layout,Bangalore South, Bangalore-560102, IndiaPhone: 91 (80) 4127-3637393, 395, 397, 399, 401, 403 Latya Road, Somdetchaopraya,Klongsan, Bangkok 10600, ThailandPhone: 66 (0) 2-861-5995 ext.123 Fax: 66 (0) 2-861-5200East Office850 / 7 Soi Lat Krabang 30 / 5, Lat Krabang Road, Lat Krabang,Bangkok 10520, ThailandPhone: 66 (0) 2-734-4434 Fax: 66 (0) 2-734-4438Jl. Jalur Sutera Blok 20A, No.16-17, Kel. Kunciran, Kec. PinangTangerang-15144, IndonesiaPhone: 62 (21) 3044-8525 Fax: 62 (21) 3044-852125, 94-Gil, Iljik-Ro, Manan-Gu, Anyang-Si, Gyeonggi-Do,13901, KoreaPhone: 82 (31) 296-7911 Fax: 82 (31) 296-7913HORIBA KOREA Ltd.KoreaRua Presbitero Plinio Alves de Souza, 645, LoteamentoMultivias, Jardim Ermida II - Jundiai Sao Paulo - CEP13.212-181 BrazilPhone: 55 (11) 2923-5400 Fax: 55 (11) 2923-5490HORIBA Instruments Brasil, Ltda.BrazilKyoto Close Moulton Park Northampton NN3 6FL UKPhone: 44 (0) 1604 542500 Fax: 44 (0) 1604 542699HORIBA UK Limited UK8F.-8, No.38, Taiyuan St. Zhubei City, Hsinchu County 30265,Taiwan (R.O.C.)Phone: 886 (3) 560-0606 Fax: 886 (3) 560-0550HORIBA Taiwan, Inc.Lot 3 and 4, 16 Floor, Detech Tower II, No.107 Nguyen Phong SacStreet, Dich Vong Hau Ward, Cau Giay District, Hanoi, VietnamPhone: 84 (24) 3795-8552 Fax: 84 (24) 3795-8553HORIBA Vietnam Company Limited Vietnam。

必修4第三单元知识点总结第一节：The Structure of the Universe1. The size of the Universe- The Universe is vast and expanding, consisting of billions of galaxies, each containing billions of stars.- It is difficult to comprehend the size of the Universe due to its immense scale.2. Galaxies- Galaxies are massive systems of stars, dust, and gas held together by gravity.- The Milky Way is the galaxy that contains our solar system.3. Stars- Stars are massive, luminous spheres of plasma that emit light and heat.- The life cycle of a star includes formation, main sequence, red giant, and white dwarf stages.4. The Big Bang Theory- The Big Bang Theory posits that the Universe began as a singular point and has been expanding ever since.- This theory has been supported by evidence such as the cosmic microwave background radiation.5. The Universe’s age and fate- The Universe is estimated to be around 13.8 billion years old.- There are different theories about the ultimate fate of the Universe, such as the Big Freeze and the Big Crunch.第二节：The Origins of the Universe1. Theories of the Universe’s origins- Theories of the Universe’s origins include the Big Bang Theory and the Steady State Theory.2. Evidence for the Big Bang- Evidence that supports the Big Bang Theory includes the redshift of distant galaxies, the cosmic microwave background radiation, and the abundance of light elements.3. The role of gravity in the formation of the Universe- Gravity played a key role in the formation of structures in the early Universe, such as galaxies and clusters of galaxies.4. The formation of light elements- Light elements like hydrogen and helium were formed in the early Universe during the process of nucleosynthesis.5. Dark matter and dark energy- Dark matter and dark energy are mysterious components of the Universe that make up a large percentage of its content.第三节：The Solar System and Planets1. The formation of the Solar System- The Solar System formed from a cloud of gas and dust called the solar nebula, with the Sun forming at its center and planets forming from the leftover material.2. Composition and structure of the Sun- The Sun is a massive ball of gas mostly composed of hydrogen and helium.- It consists of several layers, including the core, radiative zone, convective zone, photosphere, chromosphere, and corona.3. The inner planets- The inner planets of the Solar System include Mercury, Venus, Earth, and Mars.- These planets are rocky and terrestrial, with solid surfaces and relatively thin atmospheres.4. The outer planets- The outer planets of the Solar System include Jupiter, Saturn, Uranus, and Neptune.- These planets are gas giants with thick atmospheres and no solid surfaces.5. Dwarf planets and other celestial bodies- Dwarf planets like Pluto and Ceres are considered part of the Solar System, as well as other smaller objects such as asteroids, comets, and meteoroids.第四节：The Earth’s Moon1. The origin and properties of the Moon- The most widely accepted theory for the origin of the Moon is the Giant-impact hypothesis, which posits that the Moon formed from debris created by a collision between early Earth and a Mars-sized body.- The Moon is smaller and less dense than Earth and has a cratered surface with no atmosphere.2. Phases of the Moon- The Moon goes through different phases as seen from Earth, including new moon, first quarter, full moon, and last quarter, which are caused by the relative positions of the Earth, Moon, and Sun.3. Lunar eclipses and tides- Lunar eclipses occur when the Earth passes between the Sun and the Moon, causing the Earth’s shadow to fall on the Moon.- Tides on Earth are caused by the gravitational pull of the Moon, with high tides occurring on the side of the Earth facing the Moon and on the opposite side.第五节：The Earth and Its Atmosphere1. The structure of the Earth- The Earth is composed of several layers, including the inner core, outer core, mantle, and crust, with the lithosphere and asthenosphere playing key roles in the movement of tectonic plates.2. Composition and properties of the atmosphere- The Earth’s atmosphere is composed of several layers, including the troposphere, stratosphere, mesosphere, thermosphere, and exosphere.- It consists of different gases, with nitrogen and oxygen making up the majority of the atmosphere, along with trace gases and water vapor.3. The greenhouse effect- The greenhouse effect is the process by which greenhouse gases in the atmosphere trap heat from the Sun, raising the Earth’s temperature and making it suitable for life.4. Weather and climate- Weather refers to the short-term conditions of the atmosphere, including temperature, humidity, precipitation, wind, and visibility.- Climate refers to the long-term patterns of weather in a particular area.5. Natural hazards and disasters- Natural hazards such as hurricanes, tornadoes, and earthquakes can lead to natural disasters that have a significant impact on the Earth’s environment and its inhabitants.第六节：Earth’s Resources and Energy1. Renewable and nonrenewable resources- Renewable resources such as solar, wind, and hydroelectric power can be replenished naturally and are considered more sustainable than nonrenewable resources like fossil fuels.2. Fossil fuels- Fossil fuels like coal, oil, and natural gas are formed from the remains of ancient plants and animals and are a major source of energy for human civilization.3. Nuclear power- Nuclear energy is generated through the fission of uranium or plutonium atoms, producing heat that can be used to generate electricity.4. Alternative energy sources- Alternative energy sources such as solar, wind, hydroelectric, and geothermal power are being increasingly explored as cleaner and more sustainable alternatives to traditional fossil fuels.5. Energy conservation and sustainability- Energy conservation involves reducing energy usage through efficiency measures, while sustainability focuses on the long-term impact of human activity on the environment.第七节：Climate Change and Environmental Issues1. Global climate change- Climate change refers to long-term shifts in temperature and weather patterns, with global warming caused by human activities such as deforestation, burning fossil fuels, and industrialization.2. Effects of climate change- Climate change can lead to rising sea levels, extreme weather events, loss of biodiversity, and threats to human health and food security.3. Mitigation and adaptation- Mitigation involves reducing greenhouse gas emissions and transitioning to cleaner energy sources, while adaptation focuses on preparing for and addressing the impacts of climate change.4. Environmental issues- Other environmental issues such as air and water pollution, deforestation, habitat destruction, and waste management also have significant impacts on the Earth’s ecosystems and human well-being.5. Conservation and sustainable development- Conservation efforts aim to protect and preserve natural resources and wildlife, while sustainable development seeks to meet the needs of present and future generations without compromising the health of the planet.综上所述，必修4第三单元涉及了宇宙的结构与起源、太阳系与行星、地球与其大气层、地球资源与能源、气候变化与环境问题等多个方面的知识点。

a r X i v :a s t r o -p h /9607114v 1 23 J u l 1996A ROSAT survey of Hickson’s compact galaxy groupsT.J.Ponman 1,P.D.J.Bourner 1,H.Ebeling 2and H.B¨o hringer 31School of Physics and Space Research,University of Birmingham,Edgbaston,Birmingham B152TT,UK 2Institute of Astronomy,Madingley Road,Cambridge CB30HA,UK3Max-Planck-Intitut f¨u r Extraterrestrische Physik,D-85740Garching,Germany ABSTRACT We report the results of an almost complete survey of the X-ray properties of Hickson’s compact galaxy groups with the ROSAT PSPC.Diffuse X-ray emission is detected from 22groups.We infer that hot intragroup gas is present in ∼>75%of these systems and derive their X-ray luminosity function.Earlier reports that only spiral-poor systems exhibit diffuse X-ray emission are found to be incorrect.Strong correlations are found between the X-ray luminosity and both the gas temperature and the velocity dispersion of the group galaxies.We argue that these properties provide strong evidence that most of these groups are genuinely compact configurations,rather than line-of-sight parison with the X-ray properties of galaxy clusters indicate a significant steepening of the L :T relation below T ∼1keV,which may result from the action of galaxy winds.1INTRODUCTION Approximately half of all galaxies are found within groups which are probably bound (Tully1987).Whilst groups are very numerous,they are hard to identify with confidence as gravita-tionally bound systems,even in redshift catalogues,due to the problems of contamination and small number statistics which attend such poor pact galaxy groups are ones in which the galaxies are separated on the sky by only a few galactic radii.Only a small fraction of groups fall into this category,but due to their sometimes spectacular appearance,and to the comparative ease with which they can be discriminated from the field,these systems have received a good deal of attention.The most widely studied collection of compact groups are the Hickson compact groups (HCGs)compiled by Hickson (1982)by applying a set of well defined criteria for population,surface density and isolation to the Palomar Sky Survey (E band)plate collection.Numerous follow-up studies of Hickson’s catalogue of 100groups have shown that many galaxies in HCGs show morphological or kinematical peculiarities (Rubin,Hunter &Ford 1991;Mendes de Oliveira &Hickson 1994),that,relative to field galaxies,spirals tend to have low HI content (Williams &Rood 1987),and that ellipticals have low velocity dispersion for their luminosity (Zepf &Whitmore 1993).Hickson,Kindl &Huchra (1988)have also noted the existence of statistical correlations in group properties,such as a tendency for low spiral fraction to be associated with high velocity dispersion.Despite this intensive observational study,and increasing theoretical interest,the very nature of compact groups is still unclear.Numerical simulations and theoretical arguments(e.g.Barnes 1984,1985;Mamon1987;Barnes1989)indicate that,at the high densities inferred in compact groups,galaxies should merge within a few crossing times.In practice,the way in which compact groups are identified(i.e.through their high projected galaxy density)imposes strong selection effects which are likely to cause dynamical timescales to be underestimated(White1990).How-ever,these effects are hard to quantify,and Mamon(1986,1987,1995)has argued for several years that the difficulties of forming dense bound groups at the rate required to replace those lost through merging make it much more likely that the bulk of them are not truly dense in three dimensions.The observed peculiarities in the properties of HCG galaxies may still be explained if they contain real pairs,or occasionally triplets,which can interact(Mamon1992a),whilst trends such as the relationship between spiral fraction and velocity dispersion could result from redshift-dependent selection effects(Mamon1992b).A new twist has recently been added to the‘chance superposition’hypothesis by the suggestion of Hernquist,Katz&Weinberg(1995) (HKW95),that compact groups may be the result of looking along thefilamentary arrangements of galaxies which are predicted by most cosmological simulations.Meanwhile,recent numerical studies have suggested two ways in which compact groups could be continuously generated to replace those lost to the merging instability.Diaferio,Geller &Ramella(1994,1995)(hereafter DGR94,DGR95)argue that compact configurations,most of which are bound,are formed continuously during the collapse and virialisation of larger groups–though it should be noted(Mamon,private communication)that they do not include the requirement of high projected galaxy density(<26m arcsec−2)applied by Hickson,when identifying‘compact groups’in their simulated ernato,Tozzi&Cavalieri(1995) (GTC95)note that the effects of continuing infall of surrounding galaxies onto an overdense region have been ignored in earlier studies,and show that this can regenerate real compact configurations after the galaxies in the initial collapsed group have merged.Given the difficulties of small number statistics and projection effects associated with the galaxies in groups,an attractive option is to study hot gas trapped in the group potential,rather than just the galaxies themselves.In the case of galaxy clusters,this has proved a very successful approach,and the gas has turned out to constitute the bulk of the baryonic mass.The detection of significant quantities of hot gas associated with a group as a whole,would provide strong evidence for a genuine bound system,and one might hope to learn more about the dynamics and evolutionary history of groups by studying the properties of the gas in detail.Given the velocity dispersion of compact groups(∼100-400km s−1),gas temperatures∼0.1-1keV are to be expected.Hence the gas is best studied in the soft X-ray region of the spectrum. Prior to the launch of ROSAT such studies were not feasible.However,the combination of high sensitivity,low background,and reasonable spatial resolution provided by the ROSAT Position Sensitive Proportional Counter(PSPC)has changed this situation,and several studies of the X-ray properties of individual groups(Mulchaey et al1993;Ponman&Bertram1993;David et al.1994;Sulentic,Pietsch&Arp1995)and collections of groups(Ebeling,Voges&B¨o hringer 1994(EVB94);Pildis,Bregman&Evrard1995(PBE95);Saracco&Ciliegi1995(SC95);David, Jones&Forman1995;Mulchaey et al.1995;Henry et al.1995)have now been published.It is clear from these studies that some galaxy groups(both loose and compact)do contain hot X-ray emitting gas with T∼1keV,but beyond this,much of the evidence about the properties of this emission,and its correlation with other group properties,tends to be rather anecdotal, due to the lack of any sensitive survey with a reasonable level of completeness.This paper presents the results of thefirst such survey.We have combined data from the ROSAT All Sky Survey(RASS)with deeper pointings,to produce an almost complete survey of the properties of the Hickson compact galaxy groups.This allows us to study the X-ray luminosity function of these groups,and to search for relationships between the X-ray luminosity and temperature and other group properties.The data and their reduction are described in section2,and the results of the survey presented in sections3-5,after which,in section6,we consider the implications of our results for models of compact groups. Distances throughout the paper are derived assuming H0=50km s−1Mpc−1.2DATA REDUCTIONThe original catalogue of100Hickson groups(Hickson1982)was reduced by Hickson et al. (1992)to92systems,by discarding those found to contain fewer than three galaxies with accordant redshifts.In addition,we have excluded from our analysis HCG60,HCG65and HCG94,which appear to correspond to the cores of galaxy clusters(EBV94;Ramella et al.1994; Ebeling,Mendes de Oliveira&White1995),and HCG4and HCG91,which are completely dominated in the PSPC by strong X-rayflux from active galaxies within the group.HCG25and HCG54have also been excluded since they appear to be contaminated by nearby point sources or background features in the RASS data.It has become apparent(c.f.Ramella et al.1994;Rood&Struble1994;EBV94)that many compact groups are associated with larger scale structures.We have therefore retained in our study groups such as HCG5and HCG48,which may be falling into clusters,and HCG58which is apparently part of the larger galaxy group MKW10(Williams1985).Our sample therefore consists of85groups,for all of which either pointed or survey observations with the ROSAT PSPC are available.This corresponds to92%of the accordant Hickson groups. Some basic properties of these systems and of our X-ray data for them are given in Table1. Thefirst four columns,taken from Hickson et al.(1992)and Hickson(1993),show the HCG number,redshift,total number of accordant galaxies,and spiral fraction(treating S0galaxies as non-spirals).In the next column we give the line-of-sight velocity dispersion derived from the values presented by Hickson et al.(1992),which were corrected for measurement errors.We have adjusted this velocity dispersion by a factor2.1Pointed dataPointed observations were reduced in a standard way.An initial PSPC image was extracted, omitting data recorded during periods of high backgroundflux(typically a few percent of each observation)and point sources identified using a likelihood source searching program(Allan, Ponman&Jeffries,in preparation).A spectral image(i.e.an x,y,E data cube)covering the0.1-2.4keV band was then extracted for the vicinity of the group,and an estimate of the background subtracted from it.The latter was generated from a source-free region of thefield (for most groups this was taken from the annulus r=0.6−0.7◦,excluding any detected sources) and corrected to the position of the group using the known vignetting function of the instrument. Finally the background-subtracted spectral image was corrected for vignetting and divided by the exposure time,giving a map of the spectralflux.Unrelated point sources identified in thefirst stage of the procedure were removed from the image to the95%radius of the point spread function.In many groups,emission from individual galaxies was apparent.Our aim in the present survey is to study the X-ray emission from the hot intragroup gas–we therefore excludedflux from individual galaxies,except in the case of dominant ellipticals which appeared to be at the centre of the X-ray emission,since in this case it is more appropriate to identify the‘galaxy’emission with the hot gas in the core of the group potential(as seen in cD clusters).This applies in the case of the dominant galaxies HCG12a, HCG37a,HCG42a,HCG62a and HCG97a.In all other cases,galaxy emission was removed by excising a circle of radius typically1.2′.Radial profiles of theflux from individual galaxies were examined to establish that this radius was satisfactory.Except in the case of the two brightest groups,HCG62and HCG97,it was found that X-ray emission was undetectable outside a radius corresponding to200kpc.This was therefore adopted as the standard metric radius within which the signal and spectrum was evaluated for each group.In the case of HCG62and HCG97,a larger radius of500kpc was used.The detected count rate within the metric radius was extracted and compared with the Poisson error from the background.Only sources exceeding a3sigma detection threshold are regarded as having detectable diffuse emission in the following analysis.For these groups,a spectrum was extracted by collapsing the region of the spectral image cube(excluding point sources)within the metric radius.This spectrum was thenfitted with a hot plasma model(Raymond&Smith 1977)with an absorbing column frozen at the value derived from radio surveys of galactic H i (Stark et al.1992).The metallicity was generally left as a free parameter in thesefits,but was often poorly constrained.In some cases it was necessary tofix its value(0.3solar was adopted) in order to get a stablefit.EVB94noted the presence of some wisps of extremely soft emission in the vicinity of several Hickson groups,which they tentatively identified with portions of old supernova remnants within our own galaxy.We also encountered several cases in pointed data(HCG10,HCG18and HCG31) in which we recorded an extremely soft signal,confined entirely to our bottom energy channel of 0.1-0.2keV.In none of these cases was the source bright enough to image in order to determine its morphology.Due to the likelihood of contamination,we therefore treat these sources as non-detections.In the case of HCG33,we see a clear signal from the group,but this is accompanied by strongflux in the lowest energy band which is inconsistent with the high column(N H= 1.85×1021cm−2)of the source.In this case we have excluded the bottom energy channel from the spectralfitting.Thefitted spectral model has been used in each case,to derive a bolometric X-ray luminosity for detected sources.The latter were corrected for the portions masked out to remove contaminating point sources,by replacing the‘holes’in the data with interpolatedflux values.In the case of undetected sources,we have used a standard spectral model with temperature T=1keV, metallicity Z=0.3solar,and the appropriate galactic absorption column,to convert theflux limits to bolometric luminosities.The resulting parameters are listed for each group in Table2.2.2Survey dataThe RASS data at the positions of all accordant Hickson groups were analysed by EVB94using a technique,known as VTP(Ebeling&Wiedenmann1993),specially adapted to searching for extended emission.VTP gives an estimate of the sourceflux which is independent of the projected shape of the emission region.The percentage of the overallflux detected directly depends,however,on the surface brightness of the source,and corrections assuming a model profile have to be applied to recover the totalflux.For the present study we have avoided this model dependence(except for the case of HCG51–see below)by proceeding in the same way as for the above analysis of the pointed data–extractingflux estimates in the0.1-2.35keV band from a region of200kpc radius around the optical centre of each group.The background in the vicinity of each group was derived from the VTP analysis(which effec-tivelyfits a Poisson distribution to the observed distribution of counts in thefield,after removal of detected sources),and this then defines the background noise in the exposure.A3sigma detection threshold was applied,as in the pointed analysis,and a total of18accordant groups detected.However,several of these are contaminated by AGN or by probable foreground soft emission(see EVB94)or point sources,and in6cases we have higher quality pointed data. There remainfive sources(HCG51,82,83,85and86)which are detected in the RASS data,but for which no PSPC pointed data are available.In addition,we derived3sigma count rate upper limits within r=200kpc for a further48groups.The poor statistical quality of the RASS data does not permit any detailed spectral analysis,so, for both detected sources and for upper limits,we have converted count rates into bolometric luminosities using the same spectral model as for the pointed upper limits:a T=1keV,Z=0.3 Raymond&Smith plasma model,with the appropriate Stark column(given in Table2).One exception was made to the use of a metric radius of200kpc in the RASS data;HCG51is a very X-ray luminous group,andflux was detected by VTP out to a radius of at least300kpc.In this case we have therefore used the count rate derived by Ebeling et al.,which was corrected for flux missed by the VTP algorithm using an assumed King profile.This is approximately double the luminosity within r=200kpc.3SUR VEY RESULTS AND PREVIOUS WORKThe derived bolometric luminosities and spectral properties of the85groups surveyed are given in Table2.Note that the luminosities tabulated relate to diffuse emission from the groups only. Wherever emission appears to be related to a particular group galaxy,it has been removed (except in the case of dominant ellipticals)as discussed in section2.1.There are some groups for which the statistical quality of the data,or the small size of the group,did not permit emission from galaxies to be identified and removed.These systems(8in total)areflagged witha quality value q=2in Table2.Fully resolved systems are given qualityflag1,and upper limits areflagged with a‘U’.In practice,the X-ray luminosities of most of the q=2systems are sufficiently high that the contamination from galaxy emission(typically a few×1040erg s−1, judging from the resolved systems)is likely to be small.It is interesting to compare the results in Table2with those of the smaller surveys of Hickson groups carried out by PBE95and SC95.Each of these sets of authors analysed PSPC data for 12HCGs(PBE95also included the NGC2300group of Mulchaey et al.in their sample).Both found some groups to contain hot gas with T≈0.7−1.0keV.However there are significant differences between their results and ours.The most striking difference relates to low surface brightness emission.One of the main conclu-sions of PBE95was that diffuse X-ray emission is only detected in systems with spiral fractions less than50%,whereas we detect diffuseflux in several systems with high spiral content.This is a very important point,since it has a strong bearing on the questions of the reality of com-pact groups and the origin of the hot gas,as discussed in section6below.The most extreme example is HCG16,a system containing only spiral galaxies,which was studied by SC95.They concluded that there was no diffuse emission,whilst we detect emission with a low temperature (T=0.30keV),and a bolometric luminosity which is32times their quoted0.5-2.3keV upper limit!There appear to be two reasons for this dramatic difference in results from the same data.SC95 (and also PBE95)tested for the presence of diffuse emission in groups by smoothing an image and looking for signs of extension.This is not a very sensitive test for low surface brightness emission.In addition,SC95limited their data to the band0.5-2.3keV.Since wefind that spiral-rich groups with diffuse emission tend to have low temperatures,discarding the lower energy channels will have significantly reduced their sensitivity to this emission.To convince the reader that there is,in fact,extended X-ray emission in HCG16,we show in Fig.1the radial profile of the PSPC surface brightness about the centre of the optical group, after emission from the individual galaxies(and unrelated point sources)has been removed.In the present case,the emission can actually be seen in a heavily smoothed image,as is shown in Fig.2,in which an adaptive smoothing algorithm(which smoothes low surface brightness areas more heavily than high brightness regions)has been applied to a0.1-2.4keV image of the group. Finally,in Fig.3we show the PSPC spectrum of the diffuse emission(flux from the galaxies is excluded)with the bestfitting hot plasma model overlaid.In the case of PBE95,the authors appear to have been rather unlucky in their choice of groups, since we alsofind no detectable diffuse emission in their spiral-rich groups,despite our more sensitive test.In the luminous groups like HCG62,in which PBE95detect extended emission, it is noticeable that their luminosities are considerably lower than ours.This is partly due to the fact that they exclude a region around the galaxies,whilst in the case of dominant central ellipticals we include this as part of the group emission,as described in section2.1,but in addition,as they note themselves,the results are sensitive to the background level selected.In the case of HCG62,at least,they have adopted a background value which appears to be too high.An independent analysis of this group by David,Jones&Forman(1995)agrees well with our earlier results(Ponman&Bertram1993).4DISTRIBUTIONS AND CORRELATIONS4.1Luminosity and temperatureThe diffuse luminosity of our detected systems,and3σupper limits for the remainder of our sample,are plotted against their redshifts in Fig.4.Although we do not detect any of the high redshift HCGs,it is interesting to note that a group as luminous as HCG62would have been detected at any redshift out to z∼0.1.This suggests that the high redshift HCGs in our sample (remembering that we have excluded three which we believe to be cluster cores)are unlikely to be misidentified clusters.The X-ray luminosities of detected groups appear to be uncorrelated with either the number of galaxies in the group(which ranges from3to7within our detected sample)or the total optical luminosity.The lack of correlation with richness is not surprising if groups have been subject to significant merging,which would reduce the population,whilst probably increasing the amount of intragroup gas(due to galaxy winds driven by starburst activity),but the lack of any relationship between X-ray and optical luminosity,shown in Fig.5,is interesting.It strongly suggests that the hot gas is associated not with the galaxies,but with the potential well as a whole.It also indicates that the bulk of the gas did not originate from the galaxies.In Fig.6,we show the distribution of L X/L B for the detected systems(L B is taken from Hickson et al.1992,and corrected to H0=50km s−1Mpc−1).This should be compared with the typical value L X/L B≈10−4for spiral and low luminosity ellipticalfield galaxies,whilst even for very bright ellipticals the ratio barely reaches10−3(Canizares,Fabbiano&Trinchieri1987).Noting that most of the HCGs with the lowest values of L X/L B are dominated by spiral galaxies,it is apparent that the diffuse X-ray emission is too bright to be attributed to the member galaxies, even in the q=2cases where we have not been able to remove the galaxy contribution(unless previously unrecognised low luminosity active galactic nuclei are present).Approximately6%of the galaxies in Hickson groups are starburst galaxies(Zepf1993),and it is well known that such galaxies can show extended X-ray emission arising from galactic winds (e.g.Fabbiano1988).In the case of HCG16,three of the galaxies were detected by IRAS (Hickson et al.1989),and two of them(HCG16c and HCG16d,on the left in Fig.2)have the warm far infrared(FIR)colours characteristic of starburst galaxies.However,comparison with other starburst and merging systems shows that the extended emission apparent in Fig.2cannot be attributed simply to galactic winds.In normal starbursts like M82and NGC253,detectable X-ray emission from winds extends only∼10kpc from the galaxy,and even in the most luminous merging galaxies like Arp220,it is limited to a few galactic radii;whilst in HCG16,its total extent is almost∼400kpc.In addition,there is a strong correlation between FIR and X-ray luminosities,which extends from normal to merging galaxies(Ponman&Read1995).This predicts a luminosity in HCG16which agrees well with the X-ray luminiosity detected from the galaxies themselves,but is only one quarter of the observed luminosity of the whole system. The distribution of temperature for the16groups for which we have been able to extract spectra of sufficient quality tofit a hot plasma model,is shown in Fig.7.Apart from two exceptionally cool groups(HCG15and HCG16),the temperatures span only a factor of two,with most lying between0.6keV and1keV.This small range cannot be due to a temperature selection effect. For typical absorbing columns,the sensitivity of the ROSAT PSPC does not diminish greatly until T drops below0.3keV,and there is little change in count rate for a given emission measure as T rises from1to2keV.It seems that systems which look like compact groups are physically limited to a narrow range in T(apart from the cluster cores which we have rejected).Hotter systems would presumably be richer,and would fail Hickson’s isolation criterion.In order tounderstand why cooler systems are not detected in any numbers,it is helpful to look at the relationship between X-ray temperature and luminosity.The L X:T relation for our spectral subsample of16systems is shown in Fig.8.The two variables are significantly correlated:Kendall’s rank correlation coefficient(which is unit normal distributed)has a value of K=2.6(P=0.01of chance occurrence,from a two tailed test), and omitting the two q=2sources from the spectral sample,this rises to K=3.2(P=0.001).A regression line has beenfitted to the HCG points in Fig.8.It is important to allow for the errors in both luminosity and temperature when regressing,since neither are negligible,so we have used the doubly weighted regression technique discussed by Feigelson&Babu(1992),and made available through the ODRPACK package.This gives the relationship between bolometric X-ray luminosity and temperature:log L X=(43.17±0.26)+(8.2±2.7)log T.This relationship,which is very steep,is marked with its1σerror envelope,in Fig.8.In order to compare the trends in L,T and laterσ,with those seen in clusters,we have defined a set of clusters with high quality spectral data,which are also plotted on thefigure. These ten clusters are those common to the samples of Edge&Stewart(1991),from where we take the bolometric luminosity and galaxy velocity dispersion,and Yamashita(1992)who gives temperatures and metallicities derived with Ginga and(in the case of the the Perseus cluster) Tenma.The relevant parameters for these clusters are shown in Table3.A bestfit to the cluster L:T relation has recently been derived by White et al.(in preparation) by applying doubly weighted regression to a large sample of clusters.Theyfind,in our units, log L X=42.86+2.81log T,which is marked as a heavy dashed line in Fig.8.This line passes above most of the galaxy groups,and is muchflatter than the bestfit trend for the HCGs. Fitting a regression line to the sample of clusters plus HCGs shown in Fig.8,gives the relation:log L X=(42.62±0.13)+(3.29±0.17)log T,which is marked as the heavy solid line in thefigure.The steep L:T relation for groups suggests a reason for the lack of many cool groups in our sample.There is clearly a great deal of scatter about the mean trend,and the two low T groups which are detected both fall well above the regression line.Typical groups with T<0.5keV, would according to the regression,have luminosities below our detection limit of∼1041erg s−1.4.2Galaxy morphologyAs has already been noted by the authors of previous studies of the X-ray properties of sam-ples of groups(EVB94;PBE95;SC95;Mulchaey et al.1995;Henry et al.1995),there is a relationship between X-ray luminosity and galaxy morphology.In Fig.9,we show the distribu-tion of luminosity for detected groups,broken down into two classes according to whether the brightest galaxy is of early or late type.It can be seen that the brightest systems(those with L X>4×1042erg s−1)are all dominated by early-type galaxies.To assess the significance of any difference between the diffuse X-ray luminosities of HCGs with brightest Sp and E/S0galaxies,it is desirable to use not only the luminosities of detectedsystems,but also the upper limits.This can be done using‘survival analysis’,provided that the sampling is‘random’,in the sense that the limits are unrelated to the actual values of the parameter being censored.For example,this requirement would not be met in a case where longer observations are made for fainter sources,in an effort to detect them.Since we are dealing with luminosities,the actual limits are determined by both exposure time and source distance. In the case of our survey,the observation times were determined in almost all cases without any knowledge of the X-ray brightness of particular sources,and in the absence of strong evolution, the source distance is also unrelated to intrinsic luminosity.Random censoring is therefore a good assumption.We have used the ASURV package(Feigelson&Nelson1985;Isobe,Feigelson&Nelson1986), which computes the significance of the difference between two distributions using a rank statistic, which reduces to the Wilcoxon statistic in the uncensored case.Three different options are available for the weights attributed to the censored values:the Gehan,logrank,and Peto& Prentice statistics(Feigelson&Nelson1985).These tests are most reliable when the two samples being compared are comparable in size,and are censored in similar ways,both of which are true in the present case.All three statisticsfind the luminosity distributions of HCGs dominated by early and late type galaxies to differ,with significances of99%,96%and98%respectively,for the three tests.As far as the morphological mix in the groups’galaxy population is concerned,wefind only a very weak anticorrelation(K=−1.3,P=0.2)between L X and spiral fraction,f sp,in our sample,as can be seen in Fig.10.This contrasts with earlier strong statements made on the basis of smaller samples.Fig.10shows a considerably stronger relationship(K=−2.3,P=0.02) between f sp and T.Since T indicates the depth of the potential,whilst L X reflects the amount of gas trapped in it,these results suggest that galaxy morphology is related to the overall mass density in a group.4.3Velocity dispersionA fairly strong relationship between luminosity and velocity dispersion(K=1.8,P=0.07–rising to K=2.5,P=0.01if the eight q=2groups are excluded)can be seen in Fig.11. The plotted errors inσare dominated by the statistical uncertainty in determining a velocity dispersion from only a handful of galaxies.Again,we can obtain a better indication of the significance of this relationship by taking into account the censored luminosity values shown in thefiing both the Cox proportional hazard model,and the generalised Kendall rank correlation statistic,adapted for censored data as described by Isobe,Feigelson&Nelson(1986), wefind the correlation between L X andσto be significant at>99.9%confidence from either test.A2D regression on log L X and logσ,using the errors in both,giveslog L X=(30.0±5.1)+(4.9±2.1)logσ,using all22detected HCGs.As can be seen from Fig.12,this is in reasonable agreement with an extension of the trend found from clusters,shown as a dashed line:log L X=25.45+6.92logσ, (White et al.in preparation);though someflattening of the cluster trend may be indicated.A bestfit through our group+cluster sample giveslog L X=(27.6±1.4)+(5.9±0.5)logσ.。