Fabrication and evaluation of
- 格式:pdf
- 大小:1.12 MB
- 文档页数:9
《JournalofLightwaveTechnology》期刊第2页50条数据《Journal of Lightwave Technology》期刊第2页50条数据https:///doc/8514081923.html,academic-journal-foreign_journal-lightwave-technology_info_11_1/1.《Dynamic Provisioning of Self-Organized Consumer Grid Services Over Integrated OBS/WSON Networks》原文链接:https:///doc/8514081923.html,/academic-journal-foreign_journal-lightwave-technology_thesis/020417921.html2.《Experimental Demonstration of Highly Resilient Wavelength-Switched Optical Networks With a Multivendor Interoperable GMPLS Control Plane》原文链接:https:///doc/8514081923.html,/academic-journal-foreign_journal-lightwave-technology_thesis/020417922.html3.《Overlapped-Subcarrier Multiplexing for WDM Passive Optical Networks: Experimental Verification and Mathematical Analysis》原文链接:https:///doc/8514081923.html,/academic-journal-foreign_journal-lightwave-technology_thesis/020417923.html4.《A Highly Efficient Computational Model for FDTD Simulations of Ultrafast Electromagnetic Interactions With Semiconductor Media With CarrierHeating/Cooling Dynamics》原文链接:https:///doc/8514081923.html,/academic-journal-foreign_journal-lightwave-technology_thesis/020417924.html5.《Demonstration of Wireless Backhauling Over Long-Reach PONs》原文链接:https:///doc/8514081923.html,/academic-journal-foreign_journal-lightwave-technology_thesis/020417925.html6.《All Single-Mode Fiber Mach–Zehnder Interferometer Based on Two Peanut-Shape Structures》原文链接:https:///doc/8514081923.html,/academic-journal-foreign_journal-lightwave-technology_thesis/020417926.html7.《Cross-Diabolo Nanoantenna for Localizing and Enhancing Magnetic Field With Arbitrary Polarization》原文链接:https:///doc/8514081923.html,/academic-journal-foreign_journal-lightwave-technology_thesis/020418029.html8.《A 25 Gbit/s Transmitter Optical Sub-Assembly Package Employing Cost-Effective TO-CAN Materials and Processes》原文链接:https:///doc/8514081923.html,/academic-journal-foreign_journal-lightwave-technology_thesis/020418030.html9.《Reconfigurable All-Optical OTDM-to-WDM ConversionUsing a Multiwavelength Ultrashort Pulse Source Based on Raman Compression》原文链接:https:///doc/8514081923.html,/academic-journal-foreign_journal-lightwave-technology_thesis/020418031.html10.《A General Characterizing Method for Ring Resonators Based on Low Coherence Measurement》原文链接:https:///doc/8514081923.html,/academic-journal-foreign_journal-lightwave-technology_thesis/020418032.html11.《Network-Coding-Based Energy Management for Next-Generation Passive Optical Networks》原文链接:https:///doc/8514081923.html,/academic-journal-foreign_journal-lightwave-technology_thesis/020418033.html12.《Green Packet Optical Transport Networks (P-OTNs) Based on Photonic PBB-TE Switches and Minimized EEE Overhead》原文链接:https:///doc/8514081923.html,/academic-journal-foreign_journal-lightwave-technology_thesis/020418034.html13.《PMD Vector Estimation Through Time-Resolved Waveform Analysis Basedon Ultrafast xy-Field Sampling》原文链接:https:///doc/8514081923.html,/academic-technology_thesis/020418035.html14.《End-to-End Multicore Multimode Fiber Optic Link Operating up to 120 Gb/s》原文链接:https:///doc/8514081923.html,/academic-journal-foreign_journal-lightwave-technology_thesis/020418036.html15.《Modal Birefringence Analysis of Strained Buried-Core Waveguides》原文链接:https:///doc/8514081923.html,/academic-journal-foreign_journal-lightwave-technology_thesis/020418037.html16.《Investigation on the Phase Noise and EVM of Digitally Modulated Millimeter Wave Signal in WDM Optical Heterodyning System》原文链接:https:///doc/8514081923.html,/academic-journal-foreign_journal-lightwave-technology_thesis/020418038.html17.《Mode Classification and Calculation in All-Solid Photonic Bandgap Fibers》原文链接:https:///doc/8514081923.html,/academic-journal-foreign_journal-lightwave-technology_thesis/020418039.html18.《Synergetic Effects of Humidity and Temperature on PMMA Based Fiber Bragg Gratings》原文链接:https:///doc/8514081923.html,/academic-technology_thesis/020418040.html19.《Broadband Chromium-Doped Fiber Amplifiers for Next-Generation Optical Communication Systems》原文链接:https:///doc/8514081923.html,/academic-journal-foreign_journal-lightwave-technology_thesis/020418041.html20.《Resonant Fiber Optic Gyroscope Using an Air-Core Fiber》原文链接:https:///doc/8514081923.html,/academic-journal-foreign_journal-lightwave-technology_thesis/020418091.html21.《Auto Bias Control Technique Based on Asymmetric Bias Dithering for Optical QPSK Modulation》原文链接:https:///doc/8514081923.html,/academic-journal-foreign_journal-lightwave-technology_thesis/020418092.html22.《Broadband Mach–Zehnder Switch for Photonic Networks on Chip》原文链接:https:///doc/8514081923.html,/academic-journal-foreign_journal-lightwave-technology_thesis/020418093.html23.《Ferroelectric Liquid Crystal Mixture Integrated Into Optical Waveguides》原文链接:https:///doc/8514081923.html,/academic-journal-foreign_journal-lightwave-technology_thesis/020418094.html24.《Fabrication of High Glass Transition Temperature Graded-Index Plastic Optical Fiber: Part 2—Fiber Fabrication and Characterizations》原文链接:https:///doc/8514081923.html,/academic-journal-foreign_journal-lightwave-technology_thesis/020418095.html25.《Demonstration of a Remotely Dual-Pumped Long-Reach PON for Flexible Deployment》原文链接:https:///doc/8514081923.html,/academic-journal-foreign_journal-lightwave-technology_thesis/020418096.html26.《Performance Analysis for Optical OFDM Transmission in Short-Range IM/DD Systems》原文链接:https:///doc/8514081923.html,/academic-journal-foreign_journal-lightwave-technology_thesis/020418097.html27.《Simultaneous Measurement of Strain and Temperature by Using a Micro-T apered Fiber Grating》原文链接:https:///doc/8514081923.html,/academic-journal-foreign_journal-lightwave-technology_thesis/020418145.html28.《Extending the Sensing Range of Brillouin Optical Time-Domain Analysis Combining Frequency-Division Multiplexing and In-Line EDFAs》原文链journal-foreign_journal-lightwave-technology_thesis/020418146.html29.《Experimental Research of an All-Polarization-Maintaining Optical Fiber Vector Hydrophone》原文链接:https:///doc/8514081923.html,/academic-journal-foreign_journal-lightwave-technology_thesis/020418147.html30.《Photonic Crystal Fiber Interferometer for Dew Detection》原文链接:https:///doc/8514081923.html,/academic-journal-foreign_journal-lightwave-technology_thesis/020418148.html31.《Impact of Loss Variations on Double-Ended Distributed Temperature Sensors Based on Raman Anti-Stokes Signal Only》原文链接:https:///doc/8514081923.html,/academic-journal-foreign_journal-lightwave-technology_thesis/020418149.html32.《Brillouin Spectrum in LEAF and Simultaneous Temperature and Strain Measurement》原文链接:https:///doc/8514081923.html,/academic-journal-foreign_journal-lightwave-technology_thesis/020418150.html33.《Reduction in the Number of Averages Required in BOTDA Sensors Using Wavelet Denoising Techniques》原文链journal-foreign_journal-lightwave-technology_thesis/020418151.html34.《Raman-Assisted Brillouin Distributed Temperature Sensor Over 100 km Featuring 2 m Resolution and 1.2 C Uncertainty》原文链接:https:///doc/8514081923.html,/academic-journal-foreign_journal-lightwave-technology_thesis/020418152.html35.《Lab-on-a-Fiber Device for Trace Vapor TNT Explosive Detection: Comprehensive Performance Evaluation》原文链接:https:///doc/8514081923.html,/academic-journal-foreign_journal-lightwave-technology_thesis/020418153.html36.《Hybrid TDM/WDM-Based Fiber-Optic Sensor Network for Perimeter Intrusion Detection》原文链接:https:///doc/8514081923.html,/academic-journal-foreign_journal-lightwave-technology_thesis/020418154.html37.《The Use of a Fiber Comb Filter Fabricated by a CO Laser Irradiation to Improve the Resolution of a Ratiometric Wavelength Measurement System》原文链接:https:///doc/8514081923.html,/academic-journal-foreign_journal-lightwave-technology_thesis/020418155.html38.《Index Guiding Photonic Liquid Crystal Fibers forPractical Applications》原文链接:https:///doc/8514081923.html,/academic-journal-foreign_journal-lightwave-technology_thesis/020418156.html39.《Etched-Core Fiber Bragg Grating Sensors Integrated With Microfluidic Channels》原文链接:https:///doc/8514081923.html,/academic-journal-foreign_journal-lightwave-technology_thesis/020418157.html40.《Superstructure Fiber Gratings Via Single Step Femtosecond Laser Inscription》原文链接:https:///doc/8514081923.html,/academic-journal-foreign_journal-lightwave-technology_thesis/020418158.html41.《Inspection T echnique for Cleaved Optical Fiber Ends Based on Fabry–Perot Interferometer》原文链接:https:///doc/8514081923.html,/academic-journal-foreign_journal-lightwave-technology_thesis/020418159.html42.《Temperature Fiber Laser Sensor Based on a Hybrid Cavity and a Random Mirror》原文链接:https:///doc/8514081923.html,/academic-journal-foreign_journal-lightwave-technology_thesis/020418160.html43.《High-Sensitivity Coherent Optical Time DomainReflectometry Employing Frequency-Division Multiplexing》原文链接:https:///doc/8514081923.html,/academic-journal-foreign_journal-lightwave-technology_thesis/020418161.html44.《Refractive-Index Sensing With Inline Core-Cladding Intermodal Interferometer Based on Silicon Nitride Nano-Coated Photonic Crystal Fiber》原文链接:https:///doc/8514081923.html,/academic-journal-foreign_journal-lightwave-technology_thesis/020418162.html45.《Miniaturized Long-Period Fiber Grating Assisted Surface Plasmon Resonance Sensor》原文链接:https:///doc/8514081923.html,/academic-journal-foreign_journal-lightwave-technology_thesis/020418163.html46.《Polarization-Dependent In-Line Mach–Zehnder Interferometer for Discrimination of Temperature and Ambient Index Sensitivities》原文链接:https:///doc/8514081923.html,/academic-journal-foreign_journal-lightwave-technology_thesis/020418164.html47.《Long-Range Coherent OFDR With Light Source Phase Noise Compensation》原文链接:https:///doc/8514081923.html,/academic-journal-foreign_journal-lightwave-technology_thesis/020418165.html48.《Polarization Mode Coupling Involved in a Capillary Optical Fiber Sensor: Modeling and Experimental Validation》原文链接:https:///doc/8514081923.html,/academic-journal-foreign_journal-lightwave-technology_thesis/020418166.html49.《Metrological Evaluation of Optical Fiber Grating-Based Sensors: An Approach Towards the Standardization》原文链接:https:///doc/8514081923.html,/academic-journal-foreign_journal-lightwave-technology_thesis/020418167.html50.《Optical Intensity-Type Refractometer for Remote Measurements Via Fiber-Optic Cables》原文链接:https:///doc/8514081923.html,/academic-journal-foreign_journal-lightwave-technology_thesis/020418168.html。
固定修复体制作的工艺流程英文回答:Fixed Restorations Fabrication Process.1. Preoperative Evaluation.Comprehensive examination to determine the need for fixed restorations.Assessment of patient's oral health, bite relationship, and occlusion.Evaluation of existing restorations and any potential risk factors.2. Treatment Planning.Determination of the type of fixed restorations required (e.g., crowns, bridges, inlays, onlays)。
Selection of appropriate materials and designs.Estimation of treatment duration and cost.3. Tooth Preparation.Removal of decay or damaged tooth tissue.Creation of a proper shape and size for the restoration.Ensuring adequate retention and stability.4. Impression Making.Taking an impression of the prepared tooth and surrounding tissues.Using an appropriate impression material to accurately capture the details.5. Fabrication of Provisional Restoration (Optional)。
Tunnelling and Underground Space Technology 18(2003)233–2410886-7798/03/$-see front matter ᮊ2003Elsevier Science Ltd.All rights reserved.PII:S0886-7798Ž03.00032-4Evaluation ofthe load on shield tunnel lining in gravelH.Mashimo*,T.IshimuraPublic Works Research Institute,Independent Administrative Institution,Tsukuba,JapanAbstractIt is important to evaluate accurately the load acting on a shield tunnel lining to facilitate its economical and rational design.In Japan,the load calculated by Terzaghi’s formula,or overburden load,is generally adopted for the design of tunnel segments.However,some previous field measurements have shown that the actual load acting on the shield tunnel lining could be much smaller than that adopted for the design in the case of good ground conditions.In this study,field measurements at two shield tunnel construction sites in gravel were carried out,and the load acting on the shield tunnel lining was evaluated by analyzing the field data to establish the rational design of the shield tunnel segments.The method of treating the dead load of segments in the design was also investigated.ᮊ2003Elsevier Science Ltd.All rights reserved.Keywords:Shield tunnel;Lining;Load;Gravel;Dead load;Segment1.IntroductionIt is well known that segments’production cost accounts for a large part of the total shield tunnel construction cost and one of the effective methods to reduce this cost is to design the segments more efficient-ly.In the usual design method ofshield tunnel linings,firstly,the loads acting on the tunnel lining are deter-mined,and,secondary,the material and the cross-sectional dimensions ofthe segments are determined by structural calculations.It is,therefore,very important to evaluate the loads accurately.Up to now,overburden earth pressure or reduced earth pressure calculated by Terzaghi’s formula has generally been adopted as verti-cal earth pressure acting on tunnel lining for the segment design on the basis ofprevious f ield measurement data.However,some ofthe f ield measurement results have recently shown that the loads acting on the tunnel lining adopted in the design might be greater than the actual loads,particularly in case ofgood ground conditions (Koyama et al.,1995).Also the effect of the subgrade reaction,especially to the dead load ofthe segments has not been fully resolved (JSCE,1996).Therefore,it is necessary to carry out measurements and analyses of*Corresponding author.Tel.:q 81-298-796790;fax:q 81-298-796796.E-mail address:mashimo@pwri.go.jp (H.Mashimo ).the earth pressure acting on the shield tunnel lining in order to establish a rational design method for shield tunnel lining.In this study,field measurements of earth pressure,water pressure and strain in the reinforcing steel bars of the segments were carried out at two shield tunnel construction sites in gravel.The measured earth pressure and water pressure were compared with the value adopt-ed in the segment design.The measured bending moment occurring in the segments was also compared with the calculated results by using frame analysis to evaluate the load acting on the shield tunnel lining,and to determine the influence of subgrade reaction on the bending moment in the segments due to their self-weight.2.Outline of field measurementField measurements were conducted at two shield tunnel construction sites as shown in Fig. 1.The overburden height H at the measurement section of Tunnel A with a diameter D of6.2m is 9.6m,giving an overburden height to diameter ratio ofapproximately 1.5.The overburden height at the measurement section ofTunnel B with a diameter of4.75m is 12.1m,giving an overburden height to diameter ratio ofapproximately 2.5.The ring ofboth tunnels was composed ofsix reinforced concrete segments,with a thickness of 27.5234H.Mashimo,T.Ishimura /Tunnelling and Underground Space Technology 18(2003)233–241Fig.1.Description ofmeasured tunnels.Table 1Items ofmeasurements ItemsNumberofinstrument Tunnel ATunnel B Pore water pressure 4y ring =1ring 4y ring =1ring Earth pressure8y ring =1ring 8y ring =1ring Strains in reinforcing steel bars11y ring =2ring11y ring =2ringFig.2.Detail ofearth pressure cell (Tunnel B ).Fig.3.Detail ofstrain gauge attached to reinf orcing steel bar (Tunnel A ).cm and a width of100cm at Tunnel A,and a thickness of22.5cm and a width of100cm at Tunnel B.The ground at the tunnel site and its vicinity,where the measurements were carried out,appeared to be com-posed ofpermeable gravels with a standard penetrationtest value (N -value )greater than 50.Both tunnels were excavated by the earth pressure balanced shield method.Table 1shows the items ofmeasurements carried out at Tunnel A and Tunnel B.Earth pressure and water pressure acting on the tunnel lining were measured at one ring,as well as the strains in the reinforcing steel bars of the segments at two rings ofeach tunnel.The measured earth pressure is consid-ered to be the total ground earth pressure,including pore water pressure.In order to measure the earth pressure and the strains in the reinforcing steel bars,earth pressure cells of16cm in diameter at Tunnel A,and of65cm =32cm in length and breadth at Tunnel B,and strain gauges,were installed in the segments as shown in Figs.2and 3at the time off abrication.Pore water pressure gauges were mounted in grouting holes,235H.Mashimo,T.Ishimura /Tunnelling and Underground Space Technology 18(2003)233–241Fig.4.Detail ofpore water pressuregauge.Fig.5.Earth pressure and water pressure distribution (Tunnel A ).Fig.6.Earth pressure and water pressure distribution (Tunnel B ).drilled through the segments as shown in Fig.4after the backfill grouting materials were injected.3.Measurement results by earth pressure cells and pore pressure gaugesThe earth pressure measured by the earth pressure cells and the water pressure measured by the pore pressure gauges are shown in Figs.5and 6.The earth pressure and water pressure shown in the figures are those measured at the stable state approximately 3months after a segment ring was assembled,and the measured earth pressure includes the water pressure.It can be seen that the earth pressure acting on the segment reaches approximately 70kN y m at the tunnel 2crown at Tunnel A.The ground water level around the tunnel is estimated to be at the tunnel crown level according to the ground water level in a borehole in the vicinity ofthe tunnel,and the measured water pressure corresponds to the theoretical hydrostatic pressure.Con-sequently the effective overburden earth pressure P s v g H (g is the submerged unit weight )adopted as the design load reaches approximately 170kN y m and the 2water pressure is nearly zero at the tunnel crown level.The measurement data,therefore,indicate that the load acting on the tunnel lining at the tunnel crown accounts f or approximately 40–50%ofthe total amount ofthe effective overburden earth pressure and water pressure.Also it can be seen that the earth pressure measured by the earth pressure cell reaches approximately 80–170kN y m and is almost equal to the measured water 2pressure at Tunnel B.The ground water level around the tunnel estimated from the ground water level in the borehole is approximately 9m higher than the tunnel crown,and the measured water pressure corresponds to the theoretical hydrostatic pressure.Therefore it is pre-sumed that only the hydrostatic pressure acts on the tunnel lining.4.Evaluation of load on lining from measured bend-ing momentsFrame analyses by using the bedded frame model were carried out to estimate the earth pressure acting on the shield tunnel lining by comparing the calculated bending moments in the segments with measured ones.4.1.Calculation methodThe bedded frame model adopted for the calculation is shown in Fig.7.Two parallel segment rings are modeled,where each ring consists ofbeam elements representing the segments.Rotational spring elements with a rotational stiffness coefficient k represent the u segment joints which connect the segments in the236H.Mashimo,T.Ishimura /Tunnelling and Underground Space Technology 18(2003)233–241Fig.7.Bedded frame model.Fig.8.Ground conditions at the calculated section.circumferential direction,and shearing spring elements with a shearing stiffness coefficient k which represents s the ring joints which connect adjacent segment rings.The support ofthe ground that surrounds the tunnel is modeled by a continuous spring support with a coeffi-cient ofsubgrade reaction k in a normal direction,r which is not effective as the tensile spring support.The ground conditions at the calculated section are shown in Fig.8.Table 2shows the values ofthe parameters used in the calculation.In the calculation,the effective earth pressure and hydrostatic pressure,the models ofwhich are shown in Figs.9and 10,were adopted as vertical loads acting on the tunnel lining.To estimate the actual vertical load acting on the lining,calculations were carried out by using three kinds ofcombination ofvertical loads,i.e.effective overburden earth pressure together with hydrostatic pressure,effective reduced earth pressure obtained from the following Terzaghi’s formula Eq.(1)together with hydrostatic pressure,and only hydrostatic pressure.B g y c y B Ž.11P s•1y exp y K tan f •H y B (1)ŽŽ..v 01K tan f0B s R •cot p y 4q f y 2y 2Ž.Ž.1c where P s Terzaghi’s effective reduced earth pressure;v K s lateral earth pressure coefficient;c ,f s cohesion 0and angle ofinternal f riction ofsoil;g s submerged unit weight;and R s tunnel radius.c The dead load ofthe segments was not taken into237H.Mashimo,T.Ishimura /Tunnelling and Underground Space Technology 18(2003)233–241Table 2Parameters for calculationTunnelATunnel B Tunnel radius R (m )c 5.925 4.525Thickness ofsegment h (m )0.2750.225Width ofsegment w (m )1.000 1.200Moment ofinertia ofsegment I (m )40.0017730.001139Elastic modulus ofsegment E (KN y mm )2c 31.4432.36Rotational stiffness coefficient of 32.8;65.736.3;127.5segment joint k (MN Øm y rad )u Shearing stiffness coefficient of 1.961.96ring joint k (MN y m )s LoadEffective overburden earth pressure and hydrostatic pressureEffective Terzaghi’s loosening earth pressure and hydrostatic pressure Hydrostatic pressure –Self-weight of segments Lateral earth pressure coefficient l0.450.45Coefficient of ground reaction k (MN y m )3r 50(after tail leaving )50(after tail leaving )1,10,100(before tail leaving )Fig.9.Earth pressure model.Fig.10.Ground water pressure model.account in the calculation at Tunnel A,as the data collection ofstrain in the reinf orcing steel bars started after the assembly of a tunnel ring,while the dead load was taken into account at Tunnel B.The stiffness coefficients of segment joints and ring joints were determined by laboratory tests using the actual joints or theoretical calculations,and the lateral earth pressurecoefficient and the coefficient of subgrade reaction were determined from previous case studies in similar ground conditions.4.2.Calculation results and consideration4.2.1.Evaluation of loads acting on tunnel liningFig.11shows the comparison between the measured bending moments and the calculated bending moment at Tunnel A.The measured bending moments were obtained by using the measured strains in the reinforcing steel bars ofthe segments three months af ter the assem-bly ofa tunnel ring.To calculate the bending moments,238H.Mashimo,T.Ishimura /Tunnelling and Underground Space Technology 18(2003)233–241Fig.11.Bending moments distribution (Tunnel A ).Fig.12.Bending moments distribution (Tunnel B ).it is assumed that the elastic modulus ofreinf orcingsteel bars and concrete are E s 206kN y mm and E s 2s c 31kN y mm ,respectively,neglecting the effects of the 2stress ofthe concrete ifits strain is in the tensile side.The calculated results show the value for different value ofvertical load.It can be seen that the calculated results by using the effective reduced earth pressure give the closest agreement with the measured values.The effec-tive reduced earth pressure based on the bending moments accounts for approximately 60%of the effec-tive overburden earth pressure at the tunnel crown.This result is compatible with the effective earth pressure obtained from the measurement data with earth pressure cell which were approximately 40%to 50%ofthe effective overburden earth pressure.Fig.12shows the comparison between the measured bending moments and the calculated bending moments at Tunnel B.The measured bending moments were obtained in the same way as Tunnel A,assuming the elastic modulus ofconcrete E of32kN y mm and the 2c elastic modulus ofreinf orcing steel bars E of206kN ys mm .The calculated results show the value for different 2values ofvertical load.It can be seen that the calculated results by using the theoretical hydrostatic pressure give239H.Mashimo,T.Ishimura /Tunnelling and Underground Space Technology 18(2003)233–241Fig.13.Influence of subgrade reaction model.the closest agreement with the measured values.There-fore it is presumed that only the hydrostatic pressure would act on the tunnel lining.This result is also compatible with the measurements obtained from the earth pressure cells and pore pressure gauges.4.2.2.Influence of subgrade reaction modelIn the calculation as mentioned above,the subgrade reaction was basically modeled by the no-tension springs which do not resist on the tension side.However,it is thought that the evaluation ofthe ground reaction plays a very important role in the segment design.Fig.13shows the influence of the subgrade reaction model on the calculated bending moments when the subgrade reaction was modeled by linear elastic springs.It can be seen that the maximum positive and negative bending moments obtained from the calculation using linear elastic springs,which are effective as both of compres-sive and tensile springs,account for approximately 60%ofthose using no-tension springs.This indicates that the calculated results using the linear elastic springs tend to give lower bending moments than those calculated using no-tension springs.Therefore more attention should be paid to determining the value ofthe lateral earth pressure coefficient and coefficient of subgrade reaction if the subgrade reaction is modeled by linear elastic springs.4.2.3.Evaluation of ground reaction to self-weight of segmentsUp to the present,it has been thought that there was no support around a shield tunnel at the stage of assembly ofa tunnel ring in the shield machine,and that the stress occurring in the segments during assembly ofa ring remained.Theref ore in the conventional design method ofshield segments,the bending moments occur-ring due to the dead load have been calculated withouttaking account ofthe subgrade reaction.However,the bending moments due to earth pressure and water pressure have been calculated taking account ofthe ground reaction.The sum ofthese calculated bending moments has been used to determine the material and the cross-sectional dimensions ofthe segments.But according to recent experiences,it appears that little bending moment due to the dead load ofthe segments occurred at the stage ofassembling a tunnel ring,because ofthe improvements ofbackf ill grouting tech-nology,employment ofthe circle retainer and correct control ofthe jack thrust.The measured bending moments that occurred due to the dead load of segments before the tunnel ring left the tail ofshield machine in Tunnel B is shown in Fig.14.To study the influence of the subgrade reaction,the calculated bending moments due to the dead load ofthe segments with various coefficients of subgrade reaction k of1,10,100MN y m are also plotted in the figure.3r It can be seen that the bending moments that occurred due to the dead load before the tunnel ring left the tail ofshield machine are very small and the calculated bending moments taking account ofthe subgrade reac-tion with the coefficient k ofmore than 100MN y m 3r are compatible with the measured values.Furthermore,to investigate the influence of the subgrade reaction due to the dead load ofthe segments after a ring is exposed to surrounding ground together with advance ofshield machine,two kinds ofsubgrade reaction model were adopted in the calculation (see Fig.15).The standard model was the same model mentioned240H.Mashimo,T.Ishimura /Tunnelling and Underground Space Technology 18(2003)233–241Fig.14.Bending moments distribution due to dead loadofsegments.Fig.15.Model ofsubgrade reaction given by dead load ofsegments.in Section 4.1,where the subgrade reaction given by the dead load ofsegments was taken into account as the reaction ofspring support.In model I,the subgrade reaction to the dead load ofsegments was taken into account as the uniform load acting on the invert equal to the weight ofsegments.In model II,the bending241H.Mashimo,T.Ishimura /Tunnelling and Underground Space Technology 18(2003)233–241Fig.16.Influence of subgrade reaction given by dead load of segments.moment that occurred due to dead load was indepen-dently calculated without subgrade reaction.The calcu-lation results for Tunnel B are shown in Fig.16.It can be seen that the value ofthe calculated bending moments using model I is approximately two to three times larger than those measured.The calculated bending moments using model II are approximately three to four times larger than those measured.From these results,it will be desirable to take account ofthe ground reaction given by the dead load ofthe segments based on the reaction ofa spring support.5.ConclusionsIn this study,field measurements and frame analysis were carried out to evaluate the loads acting on a shield tunnel in gravel.The main results obtained from the study are as follows.1.The water pressure acting on the shield tunnel lining is almost equivalent to the theoretical hydrostatic pressure.2.When the ratio ofoverburden height H to the tunnel diameter D closes to 1.5,the vertical earth pressure acting on the shield lining is equivalent to the effec-tive earth pressure calculated by Terzaghi’s formula.However,when the ratio equals 2.5,only the hydro-static pressure acts on the shield tunnel lining,as observed by the field measurements conducted by the authors.3.At the stage ofassembling shield segments in the shield machine,small bending moments in the shield segments occurred due to their dead load.The subgra-de reaction given by the dead load ofthe segments could be taken into account by considering the reac-tion ofa spring support.ReferencesKoyama,Y.,Okano,N.,Shimizu,M.,Fujiki,I.,Y oneshima,K.,1995.In-situ measurement and consideration on shield tunnel in diluvium deposit (in Japanese ).Proceedings ofTunnel Engineering,JSCE,Vol.7.pp.385–390.Japanese Society ofCivil Engineers (JSCE ),1996.Japanese Standard for Shield Tunneling.third ed.。
No.47 Shipbuilding and Repair Quality Standard (1996) 船舶建造标准(Rev. 1, 1999)(Rev.2, Dec. 2004)(Rev.3, Nov. 2006)(Rev.4, Aug. 2008)PART A - SHIPBUILDING AND REMEDIAL QUALITY STANDARDS FOR NEW CONSTRUCTION 新建船舶建造质量标准1. Scope 范围2. General requirements for new construction 新造船舶总体要求3. Qualification of personnel and procedures 人员资质和程序认可3.1 Qualification of welders 焊工资质3.2 Qualification of welding procedures 焊接程序3.3 Qualification of NDE operators NDT人员资质4. Materials 材料4.1 Materials for structural members 结构材料4.2 Surface conditions 表面质量5. Gas Cutting 火焰切割6. Fabrication and fairness 装配和校正6.1 Flanged longitudinals and flanged brackets6.2 Built-up sections6.3 Corrugated bulkheads6.4 Pillars, brackets and stiffeners6.5 Maximum heating temperature on surface for line heating6.6 Block assembly6.7 Special sub-assembly6.8 Shape6.9 Fairness of plating between frames6.10 Fairness of plating with frames6.11 Preheating for welding hull steels at low temperature7. Alignment 对齐与对中8. Welding Joint Details焊接接头详细8.1 Typical butt weld plate edge preparation (manual welding and semi-automatic welding)8.2 Typical fillet weld plate edge preparation (manual welding and semi-automatic welding)8.3 Butt and fillet weld profile (manual welding and semi-automatic welding)8.4Typical butt weld edge preparation (Automatic welding)8.5 Distance between welds9. Remedial 修补9.1 Typical misalignment remedial9.2 Typical butt weld plate edge preparation remedial (manual welding and semi-automatic welding)9.3 Typical fillet weld plate edge preparation remedial (manual welding and semi-automatic welding)9.4 Typical fillet and butt weld profile remedial (manual welding and semi-automatic welding)9.5 Distance between welds remedial9.6 Erroneous hole remedial9.7 Remedial by insert plate9.8 Weld surface remedial9.9 Weld remedial (short bead)1. Scope 范围本标准目的是船级社认可的已有船舶建造或国家规范中未涉及的内容提供指导。
Tran s.Tian jin Univ.2010,16:388-394DOI 10.1007/s 12209-010-1414-2Accept ed d at e:2009-06-03.*S y N N S F f (N 5)LIU ,6,,D ,f LI j ,j 3@y Dyn amic Reliability Evaluation of D ouble-L ayer CylindricalLatticed Shell under Multi-Support Excitations *L IU Chung ua ng(柳春光)1,2,LI Huijun(李会军)2(1.State Key Laboratory of Coastal and Offshore Engineering,Dalian University of Technology,Dalian 116024,China;2.Faculty of Infrastructure Engineering,Dalian University of Technology,Dalian 116024,China)Tianjin University and Springer-Verlag Berlin Heidelberg 2010Ab stract :To overcome the excessive computational cost and/or bad accuracy of traditional approaches,the probabil-istic density evolution method (PDEM)is introduced.The dynamic reliability of a double-layer cylindrical latticed shell is evaluated by applying PDEM and Monte Carlo Method (MCM)respectively,and four apparent wave veloci-ties (100m/s,500m/s,800m/s and 1200m/s)and five thresholds (0.1m,0.2m,0.3m,0.4m and 0.5m)are taken into consideration.Only the difference between threshold and maximal deformation is taken as the performance func-tion.The numerical results show that results obtained by PDEM and MCM agree well;the dynamic reliability in-creases markedly with the increase of displacement threshold;the types of probabilistic density curves of response are different from that of regular distribution;the dynamic reliability will decrease with the decrease of apparent wave velocity,and more members will enter into the plastic state when subjected to multi-support excitations compared with uniform excitation.Thus,it is necessary to take the wave passage effect into consideration in the seismic design of shell structures.Keyword s :shell structure;reliability;Monte Carlo method;probability density evolution method;finite difference method;efficiencyThe large span spatial structure has been applied widely for its reasonable mechanical behavior,high rigid-ity,light weight and easy fabrication and erection.Spatial structures have been extensively used in large public and industrial buildings,such as sports and exhibition halls,theaters,terminal buildings,aero -plane hangars.Conse-quently,the seismic effects of ground motion variations on the large span spatial structure become a crucial issue.Each support of the large span spatial structure is excited asynchronously due to two important effects:(1)the time lag that seismic wave arrives at different struc-tural supports,denoted as the wave passage effect;(2)the loss of coherency of the motion due to reflections and refractions of the waves in different soil media as well as due to the diverse manners of superposition of waves arriving from an extended source at various stations,de-noted as the incoherence effect.Therefore,the analysis of seismic response of structures,especially for long span spatial structure,cannot be based on the simple assump-tion that free -field ground motions are spatially uniform.During the last three decades,the study of variation effect on the seismic response of civil structures has been ad-vanced significantly.Recent researches [1-3]have con-cluded that the difference of support motions can signifi-cantly influence the internal forces of large span spatial structure.Several response spectrum methods [4-7]have been developed to calculate the seismic response of struc-tures subjected to multi -support excitations.A great deal of researches and effort [8,9]has been directed to assess the seismic response of large span spatial structure and re-sults of these studies have shown that the spatial ground motion may induce internal forces and deformations that need to be considered in the design of reticulated shell.Nevertheless,in most cases this effect is still ignored since the seismic design codes remain unsatisfactory in terms of ground motion variations for this kind of struc-ture.This ignorance will reduce the degree of seismic safety reliability of large span spatial structure.Although much work has been done in multi -support excitations,only a few researches investigated the dynamic reliability of long span spatial structure.Especially,there have been no efficient and/or accurate approaches to evaluating theupported b ational at ural cience oundation o China o.0478094.Chunguang born in 194male r Pro .Correspondence to Hui un E-mail:lihui .LIU Chunguang et al:Dynamic Reliability Evaluation of Double-Layer Cylindrical Latticed Shell under Multi-Support Excitations—3—dynamic reliability.Hence,further studies need to be conducted to solve the problem in the dynamic reliability of large span spatial structure subjected to multi -support excitations.As for nonlinear stochastic structures,some investi-gations have been performed.For example,the Monte Carlo method (MCM),random perturbation technique and orthogonal polynomials expansion based approach have been investigated.MCM is applicable to nonlinear stochastic structural analysis,but the computational ef-forts may be exceptionally large.The random perturba-tion technique,which is more efficient in applications,is limited to the case where only small variations of random variable is feasible.Furthermore,it does not work in dy-namics problems on account of secular terms.The or-thogonal polynomials expansion based approach is lim-ited to nonlinear single degree of freedom systems with nonlinearity in the form of power series.Hence,probabil-istic density evolution method (PDEM)[10]is introduced into the dynamic reliability assessment in spatial struc-ture subjected to multi -support excitations in this paper.The numerical results show that PDEM is efficient and accurate in evaluating the dynamic reliability of spatial structure.1Multi-support excitations t o large spanspatial structure1.1Seismic analysis method for spatial structure subjected to multi-support e xcitationsStructures may be destroyed by a sudden attack of earthquake during a short period.In the past,some large span spatial structures were damaged in the earth-quake[11,12].Generally,time -history method,stochastic vibration method and response spectrum method are applied to calculating the seismic response of spatial structure sub-jected to multi -support excitations.The time -history method is easy to carry out;the advantage of stochastic vibration method is that it is constructed on the basis of statistics.Once auto -power spectrum and cross power spectrum of earthquake are determined,statistics of struc-tural response can be captured.However,it is difficult to apply the stochastic vibration to practical engineering due to its excessive computation cost and complicated mathematical derivation.The response spectrum method is only suitable for linear structures.T y ,and the commercial software can carry out the seismic analysis of spatial structure subjected to multi -support ually,three approaches are used:(1)earthquake waves exposed to structural bases by input-ting the acceleration time -history;(2)the acceleration time history integrated to the displacement time -history;(3)large mass method.In this paper,large mass method is used to carry out multi -support excitations.1.2Equation of motion of spatial structur e sub-jected to multi -support excitationsTo solve the equation of motion of spatial structure subjected to multi -support excitations,the total displace-ment are divided into pseudo -static displacement and dynamic relative displacement.The former can be ob-tained by static method,and the total displacement can be gained by substituting pseudo -static displacement into the original equation.In the absolute coordinate,nodes of spatial structures are classified into interior nodes (denoted by subscript s )and support nodes (denoted by subscript b ),and the equation of motion can be formed by partitioned matrix as follows [13,14]ss s b s s sb s s bs bb bs bb b b ++M ΜC C X XM M C C X Xss sb s bsbbbb0=K K X K K X F (1)where b X ,b X and bX are forcing displacement,velocity and acceleration vector imposed to bearing supports of latticed shell subjected to seismic excitations,respec-tively;s X ,s X and sX are displacement,velocity and ac-celeration vector imposed to non -bearing supports,re-spectively;F b is the force acting at bearing supports.Expanding the first row of Eq.(1),yieldsss s ss s ss s ++=M X C X K X sb bsb bsb bM X C X K X (2)The total displacement are divided into pseudo -static dis-placement and dynamic relative displacement,i.e.,s s d bb=+X Y Y X X 0(3)where pseudo-static displacement Y scan be obtained by assuming that the dynamic term equals zero in Eq.(1),and it yields1=Y K K K ()89he time -histor method is adopted in this paper sss s b b -4Transactions of Tianjin University V ol.16No.52010—3—When Eq.(3)is substituted into Eq.(2),it will take the formss s ss s ss s s s s ss s ++=M Y C Y K Y M Y C Y --ss ss b b s b b sb bK Y M X C X K X --(5)When Eq.(3)is substituted into Eq.(5),it yieldsss d ss d ss d ++=M Y C Y K Y ss sss ss b bsb bM Y C Y M X C X (6)Generally speaking,the damping force is propor-tional to dynamic relative velocity.Hence,s bXX can bereplaced by d Yin Eq.(1),and it yieldsss d ss d ss d ++=M Y C Y K Y ss ss b bM Y M X (7)When Eq.(4)is substituted into Eq.(7),it will take theformss d ss d ss d ++=M Y C Y K Y ss sb b+M M X ()α(8)where 1ss sb K K -=αis called pseudo-static matrix;forlumped mass matrix,sb 0=M ,then Eq.(8)takes the formss d ss d ss d ss b ++=M Y C Y K Y M X α(9)This is the equation of motion of spatial structure sub-jected to multi -support excitations.When the structure issubjected to uniform excitations,i.e.,b b b(),E u t =X Eq.(9)agrees well with the classical equation of motion.For simplification ,Eq.(9)can be rewritten as fol-lows,b()()()()t t t t ++=MY CY KY M X α(10)where b()t X is the seismic acceleration vector acting on the bearing supports of spatial structure.2D ynamic reliability of spatial structureand PDEM2.1PDEM of dynamic response of spatial struc -tureConsider the equation of motion of spatial structure subjected to random excitations as follows [15]:(,)++=MXX K X f Z ()The system has n -DOF,and M,C and K are mass,damping and stiffness matrices,respectively;f is a forc-ing function vector,and Z is a random parameter vector denoting the uncertainty of random excitation,with the known probabilistic density function p Z (Z ).The deterministic initial conditions are as follows00(),t =X x 00()t =X x (12)where X (t )is a random process determined by Z and canbe expressed as()(,)t t =X H Z (13)where H is a deterministic operator,and its component is ()(,),j j X t H t =Z 1,2,,j n = (14)The joint PDF of X (t )or the PDF of X j (t )can be deter-mined by the joint PDF of Z.The joint PDF is (,,)((,))()jj j j X p X t X H t p δ=z z z z z(15)Differentiating Eq.(15)with respect to t on bothsides,the probability density evolution equation can be derived as,,(,,)(,,)(,)0jjX j X j j p X t p X t H t t x+=Z Z z z z (16)where (,)j Ht z is the velocity of response.For t=t 0,,(,,)jX j p X t Z z can be solved with the above initial condi-tion,,0(,,)()()jX j j j p X t X X p δ=Z z z z (17)where x j ,0is the initial value of X j (t ).Finally,the PDF of ()j X t ,(,)jx j p X t ,can be ob-tained by,(,)(,,)d jjX j X z p X t p x t Ω=∫zz z(18)2.2Dynamic r elia bility asse ssmentThe dynamic reliability of response X j (t )can be ex-pressed as [16]s (){(),[0,]}j R t p X t τΩτ=∈∈(19)where p{}is the probability of event;s is the safe do-main.An absorbing boundary condition is,f (,,)0,jX j j p X t X Ω=∈Z z (failure domain )(20)Denoting the solution of initial -boundary -va lueproblem as (,,)X j p X tZ z ,the remaining PDF is(,)(,,)d j j zX j X j p X t p X t Ω=∫Zz z(21)and the reliability will be given by()(,)d j zX j jR t p X t X Ω=∫(22)90t C 11LIU Chunguang et al:Dynamic Reliability Evaluation of Double-Layer Cylindrical Latticed Shell under Multi-Support Excitations—3—2.3Procedures of dynamic re liabilityThe procedures to evaluate reliability of spatial structure subjected to multi -support excitations are as follows [15,16]:(1)discretize z in domain z and denote the lattice point as z q ,q=1,2, ,N s ,where N s is the dis-cretized points;(2)solve Eq.(11)to obtain (,)q mX t z ,where m t m t =Δ;(3)solve the initial -boundary -value problem with the finite difference method (FDM );(4)for each z q ,q=1,2, ,N s ,repeat steps (2)and (3).And then carry out the numerical integration in Eq.(18)for the dynamic response analysis,or in Eq.(22)for reliabil-ity assessment.3Dynamic reliability evaluation of lat ticed shell under m ulti-support excit ations3.1De sc ription of mode l a nd ra ndom para meters To evaluate the dynamic reliability of a large span double -layer cylindrical latticed shell subjected to multi -support excitations,PDEM is adopted.Fig.1shows the finite element model.The geometry and material proper-ties of the shell are taken as follows:span length L =108m;width B=211.2m;the height -to -span ratio is 1/3.Themembers are modeled by bar elements in the analysis.In this structure,there are 5832nodes and 6656members in all.The roof loads are taken as 2.0kN/m 2.Statistic loads and self -weight of the shell are treated as lumped masses concentrated at the joints.Initial yield stress is 240MPa,Young ’s modulus is 2.1×1011Pa and hardening modulus is 1.05×1010Pa.Joints are assumed to be hinged.The duration time of EL -Centro is 32s.(,)q mt X z in Eq.(11)is solved by ANSYS,and the initial -boundary -value problem is carried out by FDM pro-grammed by authors.Finally,the PDF of ()j X t in Eq.(18),i.e.,(,)jX j p X t and reliability assessment in Eq.(22)are also obtained by programs developed by au-thors.The probabilistic information of selected parame-ters is listed in Tab.1.Fig.1Double-layer cylindrical latticed shellTab.1Probabilistic information of random parametersRandom variablePositionMember number Distribution Mean Standard deviationE/Pa All bars1—4800Normal 2.1×10114.2×109A 1/mm 2Bars in upper-lay er,in span direction 1—858Log normal 4014.96401.5A 2/mm 2Bars in upper-lay er,in longitudinal direction 859—1722Log normal 2085.23208.52A 3/mm 2Bars in lower-lay er,in span direction 1723—2522Log normal 2968.81296.88A 4/mm 2Bars in lower-lay er,in longitudinal direction2523—3328Log normal 1545.40154.54A 5/mm 2Web member 3329—6656Log normal 1382.3138.23P/NAt upper-layer nodes1—891(node nu mber)Normal1.06P i0.074P iNote:P i (i=1—891)are t he equivalent loads concentrated at upper-layer nodes.3.2Evalua tion of dynam ic reliability of latticed shellTo demonstrate the efficiency and accuracy of PDEM in spatial structure,a large span double -layer cy-lindrical latticed shell is selected.Five displacement thresholds (0.1m,0.2m,0.3m,0.4m and 0.5m )and four apparent wave velocities (V app =100m/s,500m/s,800m/s and 1200m/s )are chosen in dynamic reliability assessment.3.2.1Dynamic reliability assessment of shell with dif-ferent thresholdsT f y y f shell is carried out considering five thresholds,i.e.,0.1m,0.2m,0.3m,0.4m and 0.5m,respectively.Two ap-parent wave velocities are considered.Only the differ-ence between threshold and maximal displacement re-sponse of node is taken as performance function.The dynamic reliabilities of latticed shell subjected to multi -support excitations are listed in Tab.2.Both PDEM and MCM are used to evaluate the dynamic reliability con-sidering two apparent wave velocities,100m/s and 500m/s.It can be seen from Tab.2that the dynamic reliabil-ity increases evidently with the increase of threshold.To fy ff y y,M M 91he evaluation o d namic reliabilit o latticed veri the e icienc and accurac C is also used.Transactions of Tianjin University V ol.16No.52010—3—Tab.2shows that results obtained by PDEM and MCM agree well whenever the threshold is high or low,and Refs.[15,16]obtained the similar solution.Furthermore,it requires only several minutes to carry out the reliability evaluation using PDEM,while MCM needs several hours,even tens of hours to achieve for a small probabil-ity event.In addition,we can see that the effect of appar-ent wave velocity on dynamic reliability of latticed shell cannot be neglected.The smaller the apparent wave ve-locity,the lower reliability is gained with the same dis-placement threshold.The time -history curves of the mean and standard deviation gained by PDEM and MCM are drawn in Figs.2and 3,when apparent wave velocities equal 100m/s and 500m/s,respectively.We can see that the introducedPDEM is accurate enough to calculate the dynamic reli-ability of latticed shell.As far as the computational ef-forts are concerned,the introduced method is more time -saving.The agreement of results between them shows the validity of the introduced algorithm.Tab.2Dynamic reliabilities of latticed shell with different thresholds and apparent wave velocities of100,500m/sReliability (100m/s)Reliability (500m/s)Threshold /mPDEM MCM PDEM MCM 0.10.2550.2560.2920.2940.20.3970.3970.4680.4620.30.6590.6580.7020.7020.40.8030.8040.8420.8460.50.9230.9250.9470.948(a )Mean (b )Standard deviationFig.2Mean and standard deviations of displacement response of latticed shell (V app =100m/s)(a )Mean (b )Standard deviationFig.3Mean and standard deviations of displacement response of latticed shell (V app =500m/s)3.2.2Dynamic reliability of latticed shell with different apparent wave velocitiesIn this section,the effect of apparent wave velocities on dynamic reliability of double-layer cylindrical latticed shell is investigated.Four apparent wave velocities are involved,i.e.,100m/s,500m/s,800m/s and 1200m/s;besides,the uniform excitation is also considered.The dynamic reliabilities are shown in Tab.2when the appar-ent wave velocities equal 100m/s and 500m/s,respec-tively.Tab.3shows the reliabilities when apparent wave velocities equal 800m/s and infinity (i.e.,uniform exci-tations ).From Tab.3we can see that the reliability in-creases evidently with the increase of displacement threshold.The values of reliability in Tab.3are larger than those in Tab.2.Moreover,the interior force of the majority of member changes evidently as the decrease of apparent wave velocity;and some values increase,while others decrease.Hence,it is necessary to take multi -92LIU Chunguang et al:Dynamic Reliability Evaluation of Double-Layer Cylindrical Latticed Shell under Multi-Support Excitations—33—support excitations into account in the dynamic reliability evaluation of long span spatial structure.Tab.3Dynamic reliabilities of latticed shell with different thresholds and ap parent wave velocities of 800m/s and infinityReliability (800m/s)Reliability(Uniform excitations)Threshold /mPDEM MCM PDEM MCM 0.10.3280.3290.3480.3490.20.5130.5140.5370.5380.30.7560.7570.7810.7830.40.8890.8890.9240.9250.50.9720.9760.9850.987Figs.4and 5show the instantaneous PDF curves of displacement response of latticed shell and their evolu-tions,considering two thresholds and two apparent wave velocities at a typical time instant.We can see that the curves are obviously different from the widely used regu-lar probabilistic distribution,e.g.,the normal distribution,log normal distribution,and the curves are irregular and sometimes have two peaks or more.Refs.[15,16]deduced the similar solution.For nonlinear response of latticed shell,the probabilistic density function curves are essen-tially different from traditional distributions despite ran-dom sources are traditional ones.Hence,the traditional second moment reliability method based on the normal distribution assumption has relatively more error.(a )V app =100m/s(b )V app =500m/sFig.4PDF curves wh en threshold is 0.4m at certain time (a )V app =100m/s(b )V app =500m/sFig.5PDF curves when threshold is 0.5m at certain time instants4Conclusions(1)The probabilistic density of response of latticedshell has evolutionary characteristics,and probabilistic density curves are different from regular distribution,e.g.,normal distribution,log normal distribution.(2)The dynamic reliability of latticed shell will de-crease with the decrease of apparent wave velocity;more members enter into the plastic state of latticed shell sub-jected to multi -support excitations compared with uni-form excitation;the dynamic reliability of spatial struc-ture subjected to multi -support excitations can be evalu-ated efficiently and accurately by PDEM,hence,the PDEM is worth further extending in spatial structure.(3)In the seismic analysis of long span spatial structure,it is necessary to take the wave passage effect into account,which has evident effect on interior force of members.Furthermore,the influence of traveling wave effect on interior force of members may be different in different places.The structural response of shell is ampli-fied by multi -support excitations.The amplification ef-fect should be considered in the seismic design of spatial structure.The amplification value changes with span and structural type of spatial structure,etc.(4)Compared with traditional approaches,PDEM y f ,9instantscan not onl obtain more probabilistic in ormation butTransactions of Tianjin University V ol.16No.52010—3—also compute efficiently and accurately.Moreover,PDEM has no limitations to secular terms and small vari-ability.The numerical results show that PDEM is accu-rate and efficient to evaluate the dynamic reliability of spatial structure subjected to multi -support excitations.Refer ences[1]Di Long,Lou Mengli.Seismic response analysis of singlelayer cylindrical reticulated shell under multi-support exci-tations[J].Journal of Tongji University (Natural Sci-ence),2006,34(10):1293-1298(in Chinese).[2]Li Zhongxian,Lin Wei,Ding Yang.Influence of wavepassage effect on seismic response of long-span spatial lat-tice structures[J].Journal of Tianjin University,2007,40(1):1-8(in Chinese).[3]Liang Jiaqing,Ye Jihong.The influence of structural spanon seismic response of space lattice structures under non-uniform excitation[J].Spatial Structures,2004,10(3):13-18(in Chinese).[4]Su Liang,Dong Shilin.Seismic analysis of two typicalspatial structures under multi-support vertical excita-tions[J].Engineering Mechanics,2007,24(2):85-90(in Chinese).[5]Dong Hexun,Ye Jihong.Effects of some parameters onpseudo-static displacement of space lattice structures under multiple support excitation[J].Journal of Southeast Uni-versity(Natural Science),2005,35(4):574-579(in Chi-nese).[6]Liang Jiaqing,Ye Jihong.Seismic response analysis oflong-span space lattice structures under multiple-support excitation [J].Journal of Southeast University (Natural Science),2003,33(5):625-630(in Chinese).[7]Ernesto Heredia-Zavoni,Vanmarcke Erik H.Seismic ran-dom-vibration analysis of multi-support-structural sys-tems[J].Journal of Engineering Mechanics,1994,120(5):1107-1128.[8]Der Kiureghian Armen,Neuenhofer Ansgar.Responsespectrum method for multi-support seismic excitations[J].Earthquake Engineering &Structural Dynamics,1992,21(8):713-740.[9]Loh Chin Hsiung,Ku Bao Ding.An efficient analysis ofstructural response for multiple-support seismic excitations [J].Engineering Structures,1995,17(1):15-26.[10]Li Jie,Chen Jianbing.Probability density evolution analy-sis of nonlinear dynamic response of stochastic struc-tures[J].A cta Mechanica Sinica,2003,35(6):716-722.[11]Fan Zhongxuan.Earthquake hazards analysis of trussstructures-roof of theater in Wu Qia of Xinjiang under earthquake 9degree [J].Earthquake Resistant Engineer-ing,1992,14(2):41-45(in Chinese).[12]Lan Tian.Seismic Design of Large Span Roof [M].ChinaArchitecture &Building Press,Beijing,2000(in Chinese).[13]Zhang Jianmin.Some Problems of Seismic Design of TrussStructures [M].Earthquake Press,Beijing,1992(in Chi-nese).[14]Chu Ye.Elastoplastic Seismic Response and StrengthFracture Analysis of Space Lattice Structures under Multi-support Excitation[D].School of Civil Engineering,Southeast University,Nanjing,2006(in Chinese).[15]Zhang Linlin,Li Jie,Peng Yongbo.Dynamic response andreliability analysis of tall buildings subject to wind loading [J].Journal of Wind Engineering and Industrial Aerody-namics,2008,96(1):25-40.[16]Li Jie,Chen Jianbing.Probability density evolutionmethod for dynamic reliability analysis of stochastic struc-tures[J].Journal of V ibration Engineering,2004,17(2):121-125.94。
Table of ContentsChapter 1. Introduction (1)1.1 Background of the Proposed Research (1)1.2 Temperature sensors (2)1.2.1 Thermocouple (2)1.2.2 Thermistors (2)1.2.3 Liquid In Glass Thermometers (3)1.2.4 Resistance Temperature Detectors (3)1.2.5 Radiation Thermometers (3)1.3 Fiber Optic Sensors (3)1.3.1 Classification of Fiber Optic Sensors (4)1.3.2 Intensity-based Fiber Optic sensor (4)1.3.3 Interferometric-based Fiber Optic sensor (5)1.3.4 Advantage of Fiber Optic Sensors (6)1.4 Scope of the Proposed Research (7)Chapter 2. Self-Calibrated Interferometric/Intensity Based (SCIIB) Sensor Technology (8)2.1 Fiber Optic Fabry-Perot Interferometer Sensors (9)2.1.1 Intrinsic Fabry-Perot Interferometer Sensor (11)2.1.2 Extrinsic Fabry-Perot Interferometer Sensor (11)2.2. SCIIB Fiber Optic Sensor System (13)2.2.1 Splitting-Spectrum Technique (15)Length (15)Coherence2.2.1.12.2.1.2 Principle of Interference (17)2.2.1.3 Principle of Self-Calibration (18)2.2.1.4 Result of Simulations (21)2.2.2 Linear Operating Range of SCIIB Sensor (23)2.2.3 Impact of Central Wavelength Drift on SCIIB Sensor System (24)2.3 Principle of SCIIB Temperature Fiber Optic Sensor (28)2.4 Advantages of the SCIIB Sensor Technology (30)Chapter 3. Design and Implementation of Multimode SCIIB Sensor System (33)3.1 Multimode Optical Fiber Sensor (33)3.2 Multimode Fiber-Based SCIIB Fiber Optic Sensor system (34)3.3 Configuration and Operation of Multimode Fiber-Based SCIIB Sensor System (34)3.4 Design and Implementation of Optical Circuit (37)3.5 Design and Implementation of Electric Circuit (39)3.5.1 Design and Implementation of Transimpedance front-end (40)3.5.2 Analysis of Noise (42)3.5.2.1 Noise in Photodiode (43)3.5.2.2 Noise in Amplifier (44)3.5.2.3 Total Voltage Noise in Transimpedance Front-end (45)3.6 Design and Implementation of Interface to Computer (45)3.6.1 Hardware of Data Acquisition System (45)3.6.2 Software of Data Acquisition System (46)3.7 Configuration of the Total Multimode Fiber-Based SCIIB Temperature Sensor System (48)Chapter 4. Design and Fabrication of Multimode SCIIB Sensor Head (49)4.1 Design of Multimode SCIIB Sensor Head (50)4.2 Sensor Head Fabrication System (51)4.3 Fabrication of SCIIB Sensor Head (52)Chapter 5. Evaluation of SCIIB Multimode Temperature Sensor System (53)5.1 SCIIB Temperature Sensor System Principle Evaluation (53)5.1.1 Self-Calibration Principle Evaluation (53)5.1.2 EFPI Temperature Sensor Principle Evaluation (54)5.2 Multimode Fiber-Based SCIIB Temperature Sensor System Performance Evaluation (55)5.2.1 Demonstration of 800 °C Multimode Fiber-Based SCIIB TemperatureSensor (55)5.2.2 Resolution of the Multimode Fiber- Based SCIIB Temperature SensorSystem (56)5.2.3 Stability of the Multimode Fiber-Based SCIIB Temperature SensorSystem (58)5.2.4 Repeatability of the Multimode Fiber-Based SCIIB Temperature SensorSystem (60)5.2.5 Accuracy of Multimode SCIIB Temperature Sensor System (61)5.2.6 Warm-up Duration of Multimode SCIIB Temperature Sensor System (62)5.3 Multimode Sensor Head Evaluation (62)5.3.1 Survivability of the Multimode SCIIB Sensor Head (62)5.3.2 Relationship between Gauge length and Operating Range of Sensor (65)Chapter 6. Conclusion (66)ReferenceVita。
NA to BS EN 1998-1:2004UK National Annex to Eurocode 8: Design of structures for earthquake resistance –Part 1: General rules, seismic actions and rules for buildingsICS 91.120.25NO COPYING WITHOUT BSI PERMISSION EXCEPT AS PERMITTED BY COPYRIGHT LAWNATIONAL ANNEXNA to BS EN 1998-1:2004Publishing and copyright informationThe BSI copyright notice displayed in this document indicates when thedocument was last issued.© BSI 2008ISBN 978 0 580 55090 4The following BSI references relate to the work on this standard:Committee reference B/525/8Draft for comment 07/30129890DCPublication historyFirst published August 2008Amendments issued since publicationAmd. no.Date Text affected© BSI 2008•i NA to BS EN 1998-1:2004ContentsIntroduction 1NA.1Scope 1NA.2Nationally Determined Parameters 2NA.3Decisions on the status of the informative annexes 11NA.4References to non-contradictory complementaryinformation 11Bibliography 12List of tablesTable NA.1 – UK values for Nationally Determined Parametersdescribed in BS EN 1998-1:2004 2Summary of pagesThis document comprises a front cover, an inside front cover,pages i andii, pages 1 to 12, an inside back cover and a back cover.NA to BS EN 1998-1:2004ii•© BSI 2008This page deliberately left blank© BSI 2008•1NA to BS EN 1998-1:2004National Annex (informative) to BS EN 1998-1:2004, Eurocode 8: Design of structures for earthquake resistance – Part 1: General rules, seismic actions and rules for buildingsIntroduction This National Annex has been prepared by BSI Subcommittee B/525/8, Structures in seismic regions . In the UK it is to be used in conjunction with BS EN 1998-1:2004.NA.1ScopeThis National Annex gives:a)the UK decisions for the Nationally Determined Parametersdescribed in the following subclauses of BS EN 1998-1:2004:b)the UK decisions on the status of BS EN 1998-1:2004 informativeannexes; andc)references to non-contradictory complementary information.2.1(1)P 5.2.2.2(10)7.1.3(4)2.1(1)P 5.2.4(1),(3)7.7.2(4)3.1.1(4) 5.4.3.5.2(1)8.3(1)3.1.2(1) 5.8.2(3)9.2.1(1)3.2.1(1),(2),(3) 5.8.2(4)9.2.2(1)3.2.1(4) 5.8.2(5)9.2.3(1)3.2.1(5) 5.11.1.3.2(3)9.2.4(1)3.2.2.1(4), 3.2.2.2(1)P 5.11.1.49.3(2)3.2.2.3(1)P 5.11.1.5(2)9.3(2)3.2.2.5(4)P 5.11.3.4(7)e)9.3(3)4.2.3.2(8) 6.1.2(1)9.3(4), Table 9.14.2.4(2)P 6.1.3(1)9.3(4), Table 9.14.2.5(5)P 6.2(3)9.5.1(5)4.3.3.1(4) 6.2(7)9.6(3)4.3.3.1(8) 6.5.5(7)9.7.2(1)4.4.2.5(2) 6.7.4(2)9.7.2(2)b)4.4.3.2(2)7.1.2(1)9.7.2(2)c)5.2.1(5)7.1.3(1),(3)9.7.2(5)10.3(2)P2•© BSI 2008NA to BS EN 1998-1:2004NA.2Nationally Determined ParametersUK decisions for the Nationally Determined Parameters described inBS EN 1998-1:2004 are given in Table NA.1.Table NA.1UK values for Nationally Determined Parameters described in BS EN 1998-1:2004Subclause Nationally Determined Parameter Eurocode recommendation UK decision2.1(1)P Reference return period T NCR of seismic action for the no-collapse requirement (or, equivalently , reference probability of exceedance in 50 years, P NCR ).T NCR =475 years P NCR =10%In the absence of a project-specific assessment,adopt a return periodT NCR of 2 500 years. Furtherguidance is given in PD 6698.2.1(1)P Reference return period T DLR of seismic action for the damage limitation requirement (or, equivalently , reference probability of exceedance in 10 years, P DLR ).T DLR =95 years P DLR =10%In the absence of a project-specific assessment,adopt the recommended values.Further guidance is given in PD 6698.3.1.1(4)Conditions under which ground investigations additional to those necessary for design for non-seismic actions may be omitted and default ground classification may be used.[None]The need for additional ground investigationsshould be established on a site-specific basis. Further guidance is given in PD 6698.3.1.2(1)Ground classification scheme accounting for deep geology , including values of parameters S , T B , T Cand T D defining horizontal and vertical elastic response spectra in accordance with BS EN 1998-1:2004, 3.2.2.2 and 3.2.2.3.[None]There is no requirement to account for deep geology .Further guidance is given in PD 6698.3.2.1(1),(2),(3)Seismic zone maps and reference ground accelerations therein.[None]In the absence of a project-specific assessment, adopt the reference ground accelerations for a return period T NCR of 2 500 years given by theseismic contour map in PD 6698.3.2.1(4)Governing parameter (identification and value) for threshold of low seismicity .a g u 0,78m/s 2 or a g S u 0,98m/s 2a g u 2m/s 2 (for T NCR =2 500 years)3.2.1(5)Governing parameter (identification and value) for threshold of very low seismicity .a g u 0,39m/s 2 or a g S u 0,49m/s 2a g u 1.8 m/s 2 (for T NCR =2 500 years)© BSI 2008•3NA to BS EN 1998-1:2004Table NA.1UK values for Nationally Determined Parameters described in BS EN 1998-1:2004 (continued )SubclauseNationally Determined Parameter Eurocode recommendation UK decision3.2.2.1(4), 3.2.2.2(1)P Parameters S, T B , T C , T D defining shape of horizontal elastic response spectra.In the absence of deep geology effects, and for Type 1 spectra (where earthquakes that contribute most to the seismic hazard defined for the site for the purpose of probabilistic hazard assessment have a surface-wave magnitude, M s , greater than 5,5):In the absence of site-specific information, therecommended values for Type 2 earthquakes maybe used, but see also PD 6698.Ground type S T B (s)T C (s)T D (s)A 1,00,150,42,0B 1,20,150,52,0C 1,150,200,62,0D 1,350,200,82,0E 1,40,150,52,0In the absence of deep geology effects, and for Type 2 spectra (where earthquakes that contribute most to the seismic hazard defined for the site for the purpose of probabilistic hazard assessment have a surface-wave magnitude, M s , less than 5,5):Ground type S T B (s)T C (s)T D (s)A 1,00,050,251,2B 1,350,050,251,2C 1,50,100,251,2D 1,80,100,301,2E 1,60,050,251,23.2.2.3(1)P Parameters a vg , T B , T C , T D defining shape of vertical elastic response spectra.Spectrum a vg /a g T B (s)T C (s)T D (s)In the absence of site-specific information, therecommended values for Type 2 earthquakes maybe used, but see also PD 6698.Type 10,900,050,151,0Type 20,450,050,151,03.2.2.5(4)P Lower bound factor β on design spectral values. 0,2Use the recommended value.4•© BSI 2008NA to BS EN 1998-1:2004Table NA.1UK values for Nationally Determined Parameters described in BS EN 1998-1:2004 (continued )Subclause Nationally Determined Parameter Eurocode recommendation UK decision4.2.3.2(8)Reference to definitions of centre of stiffness and of torsional radius in multi-storey buildings meeting or not conditions (a) and (b) of BS EN 1998-1:2004, 4.2.3.2(8).[None]Any appropriate method may be used.Further guidance is given in PD 6698.4.2.4(2)P Ratio ϕ of coefficient ψEi on variable mass used in seismic analysis to combination coefficient ψ2i for quasi permanent values of variable actions.Type of variable action Storey ÎUse the recommended values. Storeys occupied by different tenants may be considered asindependently occupied.Categories A–C*Roof 1,0Storeys with correlated occupancies0,8Independently occupied storeys 0,5Categories D–F*and Archives 1,0* Categories as defined in BS EN 1991-1-1:2002.4.2.5(5)P Importance factor γI for buildings.Class I:γI =0,8Class III:γI =1,2Class IV:γI =1,4Where a value for the reference returnperiod T NCR of 2 500 years has been adoptedfor CC3 structures, γI =1 should be assumed.Where T NCR has been assessed on aproject-specific basis, γI should also be chosenon a project-specific basis. Further guidance is given in PD 6698.4.3.3.1(4)Decision on whether nonlinear methods of analysis may be applied for the design of non-base-isolated buildings. Reference to information on member deformation capacities and the associated partial factors for the Ultimate Limit State for design or evaluation on the basis of nonlinear analysis methods.[None]No supplementary advice.© BSI 2008•5NA to BS EN 1998-1:2004Table NA.1UK values for Nationally Determined Parameters described in BS EN 1998-1:2004 (continued )Subclause Nationally Determined Parameter Eurocode recommendation UK decision4.3.3.1(8)Threshold value of importance factor, γI , relating to the permitted use of analysis with two planar models.[None]3D (spatial) analysis models are recommended forall consequence class CC3 buildings.4.4.2.5(2)Overstrength factor γRd for diaphragms.For brittle failure modes, such as shear, γRd =1,3.For ductile failure modes, γRd=1,e the recommended values.4.4.3.2(2)Reduction factor ν for displacements at damage limitation limit state.Class I & II:É =0,4Class III & IV:É =0,5In consequence class CC3 buildings, storey drifts should be checked against the specified limits using the recommended values of reduction factorν.5.2.1(5)Geographical limitations on use of ductility classes for concrete buildings.[None]There are no geographical limitations.5.2.2.2(10)q o -value for concrete buildings subjected to special Quality System Plan.Adjustment to q o -value is a factor in the range 1 to 1,2, with no recommended value within this range.An adjustment factor of up to 1,2 onq o ispermitted if a formal quality plan is applied to the design, procurement and construction. The design quality plan should include a peer review of the seismic design and the construction quality plan should include special inspection measures for the critical (dissipative) regions.5.2.4(1), (3)Material partial factors for concrete buildings in the seismic design e the γc and γs values for the persistent and transient design e the recommended values.5.4.3.5.2(1)Minimum web reinforcement of large lightly reinforced concrete walls.The minimum value for walls given in BS EN 1992-1-1:2002 and its National e the recommended values.5.8.2(3)Minimum cross-sectional width b w, min and depth h w, min of concrete foundation beams. Buildings up to 3 storeys:bw, min =0,25mh w, min =0,4mBuildings with 4 or more storeys:b w, min =0,25mh w, min =0,5mUse the recommended values.6•© BSI 2008NA to BS EN 1998-1:2004Table NA.1UK values for Nationally Determined Parameters described in BS EN 1998-1:2004 (continued )SubclauseNationally Determined Parameter Eurocode recommendation UK decision5.8.2(4)Minimum thickness t min and reinforcement ratioρs, min of concrete foundation slabs.t min = 0,2mρs, min = 0,2%Use the recommended values.5.8.2(5)Minimum reinforcement ratio ρb, min of concrete foundation beams.ρb, min = 0,4%ρb, min = 0,2% in top face and 0,2% in bottom face.5.11.1.3.2(3)Ductility class of precast wall panel systems.DCM Use the recommended value.5.11.1.4Factor k p on q -factors of precast systems.k p = 1,0 for structures with connections conforming to BS EN 1998-1:2004, 5.11.2.1.1,5.11.2.1.2, or 5.11.2.1.3k p = 0,5 for structures with other types ofconnection Use the recommended values.5.11.1.5(2)Ratio A p of transient seismic action assumed during erection of precast structures to design seismic action defined in BS EN 1998-1:2004, Section 3.A p = 0,3 unless otherwise specified by special studies In the absence of a site-specific assessment, use therecommended value.5.11.3.4(7)e)Minimum longitudinal steel ρc, min in grouted connections.ρc, min =1%Use the recommended value.6.1.2(1)Upper limit of q for low-dissipative structural behaviour concept.1,52Further guidance is given in PD 6698.Limitations on structural behaviour concept.[None]No limitations on structural behaviour concept. Further guidance is given in PD 6698.Geographical limitations on use of ductility classes for steel buildings.[None]No geographical limitations. Further guidance is given in PD 6698.6.1.3(1)Material partial factors for steel buildings in the seismic design e the γs values for the persistent and transient design e the recommended values.6.2(3)Overstrength factor for capacity design of steel buildings.γov = 1,25Use the recommended value.© BSI 2008•7NA to BS EN 1998-1:2004Table NA.1UK values for Nationally Determined Parameters described in BS EN 1998-1:2004 (continued )SubclauseNationally Determined Parameter Eurocode recommendation UK decision6.2(7)Information as to how BS EN 1993-1-10:2005 – selection of steel for fracture toughness and through thickness properties – may be used in the seismic design situation.[None]The fracture toughness and through thicknessproperties of the steel should be selected on a project-specific basis. Further guidance is given in PD 6698.6.5.5(7)Reference to complementary rules on acceptable connection design.[None]Complementary rules for connection design may be developed on a project-specific basis. Further guidance is given in PD 6698.6.7.4(2)Residual post-buckling resistance of compression diagonals in steel frames with V -bracings.γpb = 0,3γpb = γpb * N b,Rd (λbar)/ Npl,Rd(γpb * times design buckling resistance over plasticresistance)γpb * = 0,7 for q u 2= 0,3 for qW 5For 2u q u 5, γpb *= 0,3 may be assumed or refer to PD 6698.Further guidance is given in PD 6698.7.1.2(1)Upper limit of q for low-dissipative structural behaviour concept.1,52Limitations on structural behaviour concept.[None]No limitations on structural behaviour concept.Geographical limitations on use of ductility classes for composite steel-concrete buildings.[None]No geographical limitations.7.1.3(1),(3)Material partial factors for composite steel-concrete buildings in the seismic design e the γs values for the persistent and transient design e the recommended values.7.1.3(4)Overstrength factor for capacity design of composite steel-concrete buildings.γov = 1,25Use the recommended value.8•© BSI 2008NA to BS EN 1998-1:2004Table NA.1UK values for Nationally Determined Parameters described in BS EN 1998-1:2004 (continued )SubclauseNationally Determined Parameter Eurocode recommendation UK decision7.7.2(4)Stiffness reduction factor for concrete part of a composite steel-concrete column section.r =0,5In the absence of special studies, use the recommended value.8.3(1)Geographical limits on ductility class for timber buildings.[None]No geographical limits.9.2.1(1)Type of masonry units with sufficient robustness.[None]Any type of masonry unit listed in BS EN 1996-1-1:2005, Table 3.1, is acceptable.9.2.2(1)Minimum strength of masonry units.f b,min = 5N/mm 2 (normal to bedface)f bh,min = 2N/mm 2 (parallel to bedface)Use the minimum values given inBS EN 1996-1-1:2005.9.2.3(1)Minimum strength of mortar in masonry buildings.f m,min = 5N/mm 2 (unreinforced or confined masonry)f m,min = 10N/mm 2 (reinforced masonry)Use the minimum values given inBS EN 1996-1-1:2005.9.2.4(1)Alternative classes for perpend joints in masonry . [None]Perpend joints fully grouted with mortar orungrouted joints with mechanical interlocking between masonry units may be used. Ungrouted joints without mechanical interlock may only be used subject to appropriate validation.9.3(2)Conditions for use of unreinforced masonry satisfying provisions of BS EN 1996-1:2005 alone.[None]There are no restrictions on the use of unreinforced masonry that follows the provisions of BS EN 1996-1:2005 alone.9.3(2)Minimum effective thickness t ef,min of unreinforced masonry walls satisfying provisions of BS EN 1996-1:2005 alone.t ef,min = 240 mm t ef,min = 170 mm in cases of low seismicityt ef,min = 170 mm9.3(3)Maximum value of ground acceleration a g,urm for the use of unreinforced masonry satisfying provisions of BS EN1998-1.a g,urm = 0,2 g a g,urm = 0,25 g9.3(4), Table 9.1q -factor values in masonry buildings.Unreinforced masonry in accordance with BS EN 1998-1: q = 1,5Confined masonry: q = 2,0Reinforced masonry: q = 2,5Unreinforced masonry inaccordance with BS EN 1998-1:q = 2,0Confined masonry: q = 2,5Reinforced masonry: q = 3,0© BSI 2008•9NA to BS EN 1998-1:2004Table NA.1UK values for Nationally Determined Parameters described in BS EN 1998-1:2004 (continued )SubclauseNationally Determined Parameter Eurocode recommendation UK decision9.3(4), Table 9.1q -factors for buildings with masonry systems which provide enhanced ductility .[None]Enhanced values need to be justified on a case-by-case basis.9.5.1(5)Geometric requirements for masonry shear walls.Masonry type t ef,min (mm)(h ef /t ef )max (l /h )min Use the recommended values.Unreinforced, with natural stone units 35090,5Unreinforced, with any other type of units240120,4Unreinforced, with any other type of units, in cases of low seismicity170150,35Confined masonry 240150,3Reinforced masonry 24015No restrictionSymbols used have the following meaning:t ef thickness of the wall (seeBS EN 1996-1-1:2005);h efeffective height of the wall (see BS EN 1996 1-1:2005);h greater clear height of the openings adjacent to the wall;l length of the wall.9.6(3)Material partial factors in masonry buildings in the seismic design situation.γm = 2/3 of value specified in National Annex to BS EN 1996-1-1:2005, but not less than 1,5γs = 1,0Use the recommended values.10•© BSI 2008NA to BS EN 1998-1:2004Table NA.1UK values for Nationally Determined Parameters described in BS EN 1998-1:2004 (continued )SubclauseNationally Determined Parameter Eurocode recommendation UK decision9.7.2(1)Maximum number of storeys and minimum area of shear walls of “simple masonry building”.Acceleration at site a g .S u0,07k ⋅g u 0,10k ⋅g u 0,15k ⋅g u 0,20k ⋅g Use the recommended values, unless justified on a project-specific basis.Further guidance is given in PD 6698.Type of construction Number of storeys (n )**Minimum sum ofcross-sections areas ofhorizontal shear walls in each direction, as percentage of the total floor area per storey (p A,min )Unreinforced masonry 12342,0%2,0%3,0%5,0% 2,0%2,5%5,0%n/a*3,5%5,0%n/a n/a n/an/an/a n/aConfined masonry 23452,0%2,0%4,0%6,0%2,5%3,0%5,0%n/a 3,0%4,0%n/a n/a 3,5%n/an/a n/aReinforced masonry 23452,0%2,0%3,0%4,0%2,0%2,0%4,0%5,0%2,0%3,0%5,0%n/a 3,5%5,0%n/a n/a* n/a means “not acceptable”.** Roof space above full storeys is not included in the number of storeys.9.7.2(2)b)Minimum aspect ratio in plan λmin of “simple masonry buildings”.λmin = 0,25Use the recommended value.9.7.2(2)c)Maximum floor area of recesses in plan for “simple masonry buildings”, expressed as a percentage p maxof the total floor plan area above the level considered.p max = 15%Use the recommended value.9.7.2(5)Maximum difference in mass Δm, max and wall area ΔA, max between adjacent storeys of “simple masonry buildings”.Δm, max= 20%ΔA, max= 20%Use the recommended values.10.3(2)P Magnification factor γx on seismic displacements for isolation devices.γx = 1,2 for buildings γx = 1,5 for buildingsNA to BS EN 1998-1:2004 NA.3Decisions on the status of theinformative annexesNA.3.1Elastic displacement response spectrum[BS EN 1998-1:2004, Annex A]BS EN 1998-1:2004 informative Annex A should not be used in the UK.Further guidance is given in PD 6698.NA.3.2Determination of the target displacement for nonlinear static (pushover) analysis[BS EN 1998-1:2004, Annex B]BS EN 1998-1:2004 informative Annex B may be used in the UK as aninformative annex. Further guidance is given in PD 6698.NA.4References to non-contradictorycomplementary informationThe following is a list of references that contain non-contradictorycomplementary information for use with BS EN 1998-1:2004.•PD 6698:2008, Background paper to the UK National Annexes to BS EN 1998-1, BS EN 1998-2, BS EN 1998-4, BS EN 1998-5and BS EN 1998-6;•Manual for the seismic design of steel and concrete buildingsto Eurocode 8. Institution of Structural Engineers, London. Indraft; publication expected 2008.© BSI 2008•11NA to BS EN 1998-1:200412•© BSI 2008BibliographyStandards publicationsBS EN 1993-1-10:2005, Eurocode 3 – Design of steel structures –Part 1-10: Material toughness and through-thickness properties BS EN 1996-1-1:2005, Eurocode 6 – Design of masonry structures –Part 1-1: General rules for reinforced and unreinforced masonry structuresBS EN 1998-1:2004, Eurocode 8 – Design of structures for earthquake resistance – Part 1: General rules, seismic actions and rules for buildingsPD 6698:2008, Background paper to the UK National Annexesto BS EN 1998-1, BS EN 1998-2, BS EN 1998-4, BS EN 1998-5 andBS EN 1998-6Other publications[1] Institution of Structural Engineers: Manual for the seismic design of steel and concrete buildings to Eurocode 8, London: publication expected 2008.NA to BS EN 1998-1:2004 This page deliberately left blankBSI – British Standards Institution BSI is the independent national body responsible for preparing British Standards. It presents the UK view on standards in Europe and at the international level. It is incorporated by Royal Charter.RevisionsBritish Standards are updated by amendment or revision. Users of British Standards should make sure that they possess the latest amendments or editions.It is the constant aim of BSI to improve the quality of our products and services. We would be grateful if anyone finding an inaccuracy or ambiguity while using this British Standard would inform the Secretary of the technical committee responsible, the identity of which can be found on the inside front cover.Tel:+44(0)2089969000 Fax:+44(0)2089967400BSI offers members an individual updating service called PLUS which ensures that subscribers automatically receive the latest editions of standards.Buying standardsOrders for all BSI, international and foreign standards publications should be addressed to Customer Services. Tel:+44(0)2089969001Fax:+44(0)2089967001 Email:orders@You may also buy directly using a debit/credit card from the BSI Shop on the Website /shop.In response to orders for international standards, it is BSI policy to supply the BSI implementation of those that have been published as British Standards, unless otherwise requested.Information on standardsBSI provides a wide range of information on national, European and international standards through its Library and its Technical Help to Exporters Service. Various BSI electronic information services are also available which give details on all its products and services. Contact the Information Centre. Tel:+44(0)2089967111 Fax:+44(0)2089967048 Email:info@Subscribing members of BSI are kept up to date with standards developments and receive substantial discounts on the purchase price of standards. For details of these and other benefits contact Membership Administration. Tel:+44(0)2089967002 Fax:+44(0)2089967001 Email:membership@Information regarding online access to British Standards via British Standards Online can be found at /BSOL.Further information about BSI is available on the BSI website at.CopyrightCopyright subsists in all BSI publications. BSI also holds the copyright, in the UK, of the publications of the international standardization bodies. Except as permitted under the Copyright, Designs and Patents Act 1988 no extract may be reproduced, stored in a retrieval system or transmitted in any form or by any means – electronic, photocopying, recording or otherwise – without prior written permission from BSI.This does not preclude the free use, in the course of implementing the standard, of necessary details such as symbols, and size, type or grade designations. If these details are to be used for any other purpose than implementation then the prior written permission of BSI must be obtained.Details and advice can be obtained from the Copyright & Licensing Manager. Tel:+44(0)2089967070 Email:copyright@ NA to BS EN 1998-1:2004BSI Group Headquarters 389 Chiswick High Road,London W4 4AL, UK Tel +44 (0)20 8996 9001Fax +44 (0)20 8996 7001/standards。
d e n t a l m a t e r i a l s24(2008)235–243Fabrication and evaluation of Bis-GMA/TEGDMA dentalresins/composites containing nanofibrillar silicateMing Tian a,b,Yi Gao a,Yi Liu a,Yiliang Liao a,Nyle E.Hedin a,Hao Fong a,∗a Department of Chemistry,South Dakota School of Mines and Technology,Rapid City,SD57701-3995,United Statesb The Key Laboratory of Beijing City on Preparation and Processing of Novel Polymer Materials,Beijing University of Chemical Technology,Beijing100029,Chinaa r t i c l e i n f oArticle history:Received20January2007Received in revised form24March2007Accepted7May2007Keywords:Dental materialNanofibrillar silicateBis-GMATEGDMAa b s t r a c tObjective.To investigate the reinforcement of Bis-GMA/TEGDMA dental resins(without con-ventional glassfiller)and composites(with conventional glassfiller)with various massfractions of nanofibrillar silicate(FS).Methods.Three dispersion methods were studied to separate the silanized FS as nano-scaled single crystals and uniformly distribute them into dental matrices.The photo-curingbehaviors of the Bis-GMA/TEGDMA/FS resins were monitored in situ by RT-NIR to studythe photopolymerization rate and the vinyl double bond conversion.Mechanical proper-ties(flexural strength,elastic modulus and work-of-fracture)of the nano FS reinforcedresins/composites were tested,and analysis of variance(ANOVA)was used for the statisticalanalysis of the acquired data.The morphology of nano FS and the representative fracturesurfaces of its reinforced resins/composites were examined by SEM/TEM.Results.Impregnation of small mass fractions(1%and 2.5%)of nano FS into Bis-GMA/TEGDMA(50/50mass ratio)dental resins/composites improved the mechanicalproperties rger mass fraction of impregnation(7.5%),however,did not fur-ther improve the mechanical properties(one way ANOVA,P>0.05)and may even reduce themechanical properties.The high degree of separation and uniform distribution of nanoFS into dental resins/composites was a challenge.Impregnation of nano FS into dentalresins/composites could result in two opposite effects:a reinforcing effect due to the highlyseparated and uniformly distributed nano FS single crystals,or a weakening effect due tothe formation of FS agglomerates/particles.Significance.Uniform distribution of highly separated nano FS single crystals into den-tal resins/composites could significantly improve the mechanical properties of theresins/composites.©2007Academy of Dental Materials.Published by Elsevier Ltd.All rights reserved.1.IntroductionDental composites consisting of polymeric resin matrices andinorganicfillers have been available for over four decades.Compared to dental amalgams,the composites possess betteresthetic property,have less safety concern,and have shown∗Corresponding author.Tel.:+16053941229;fax:+16053941232.E-mail address:Hao.Fong@(H.Fong).reasonably satisfactory clinic results.They have been widelyadopted by the dental profession as the restorative materialof choice.Dental resins are usually cured(hardened)by photo-initiated free radical polymerization.Camphorquinone(CQ)is a commonly used visible-light initiator and ethyl-4-(N,N -0109-5641/$–see front matter©2007Academy of Dental Materials.Published by Elsevier Ltd.All rights reserved.doi:10.1016/j.dental.2007.05.002236d e n t a l m a t e r i a l s24(2008)235–243dimethylamino)benzoate(4EDMAB)is a commonly used co-initiator.The monomer2,2 -bis-[4-(methacryloxypropoxy)-phenyl]-propane(Bis-GMA)has been widely used as an important dental base monomer since it was invented in early1960s[1,2].Bis-GMA is a very viscous liquid.To improve the handling qualities,a low viscosity diluent monomer,such as tri(ethylene glycol)dimethacrylate(TEGDMA),is added to thin the resin.In Bis-GMA/TEGDMA dental resin,Bis-GMA functions to limit the photopolymerization induced volumet-ric shrinkage and to enhance resin reactivity,while TEGDMA provides for the increased vinyl double bond conversion [3,4].While dental resins have been reinforced with inorganic fillers such as silanized glass/ceramic particles for years,the relatively low strength and durability of the composites have limited their uses[5–8].The dental composites haveflexural strength in the range from100to140MPa,which can fulfill the requirements of small restorations but cannot survive large stress-bearing restorations.Furthermore,the strength of den-tal composites decreases significantly after long-term water aging.The average lifetime of dental composites is less than 5years[5].In comparison,dental amalgams have a lifetime of more than15years[9,10].Investigations of the failures revealed that,among other things,the inorganicfiller was a major contributor to the failures[11,12].Many inorganicfillers currently being used for dental composites are spherical or irregular in shape.Suchfiller particles at occlusal surfaces are susceptible to dislodgement from the resin matrix during wear with food boluses.This causes the reinforcing effect to be lost.Reinforcement with high-strengthfibers/whiskers has been shown to result in dramatic improvements on the properties of dental composites[13–16].Research conducted by Xu and co-workers revealed that the impregnation of extremely strong ceramicfibers/whiskers could lead to a two-fold increase in composite strength and toughness,as well as provide promising results in composite polishability,water absorption and strength durability[17–21].The“bridging”mechanism has been proposed to explain the reinforcement by thefibers/whiskers.If a micro-crack is initiated in a matrix under contact wear and/or other stresses,thefibrillarfillers remain intact across the crack planes and support the applied load.Crack-opening is therefore resisted by the bridgingfillers and the matrix is reinforced.Requirements for thefillers to achieve the effective“bridging”reinforcement include high mechanical properties(especially strength and modulus)and a large aspect ratio.Unlike micron-scaledfibers/whiskers,fibrillar silicate(FS) is composed of nano-scaled single crystals(fibers).FS is a class of hydrated magnesium/aluminum silicate,and there are several types of FS minerals found in nature.The most abundant type is known as attapulgite/palygorskite,which is found mostly in the United States and China.The FS used in this study was attapulgite obtained from China,and its chem-ical formula is Mg5[Al]Si8O20(HO)2(OH2)4·4H2O.The primary structural units of FS are the silicate single crystals that are 100–3000nm in length and10–25nm in diameter,and these single crystals stack/agglomerate into particles with sizes in microns[22,23].The FS nano-scaled single crystals possess a high degree of structural perfection and superior mechanical properties.For example,the strength of a FS nano-scaled sin-gle crystal is over50GPa[22],which is at least10times higher than that of most micron-scaledfibers/whiskers.Unlike lay-ered silicates such as montmorillonite,which are difficult to completely exfoliate into nano-scaled silicate layers and to uniformly distribute in dental matrices,FS is relatively easy to separate into nano-scaled single crystals and to distribute uniformly in dental matrices.This is because the spacing among the aggregated single crystals in FS is much larger than that of the silicate layers in montmorillonite.As a result, the interaction of the single crystals in FS is considerably weaker than that of the silicate layers in montmorillonite. Therefore,without chemical substitution of metal ions with surfactants such as tertiary amine ions(a widely adopted method for intercalation/exfoliation of montmorillonite to prepare nanocomposites),FS can be readily separated into nano-scaled single crystals by simply dispersing FS agglom-erates/particles in polar solvents like ethanol,followed by vigorously mechanical stirring[23].Additionally,the inter-facial bonding between the silanized FS nano-scaled single crystalfiller and the dental resin matrix can be reasonably strong since there are abundant Si–OH groups on the surface of FS single crystals,and these groups can react with silane coupling agents such as3-methacryloxypropyltrimethoxy (MPTMS).The aim of this study was to investigate the reinforcements of Bis-GMA/TEGDMA dental resins(without conventional glassfiller)and composites(with conventional glassfiller) with various mass fractions of nano FS.The hypothesis was that uniform distribution of the silanized FS nano-scaled single crystals(fibers)into Bis-GMA/TEGDMA dental resins/composites would result in substantial improvements of the mechanical properties.To test this hypothesis,photo-cured Bis-GMA/TEGDMA dental resins/compositesfilled with various mass fractions of silanized nano FS were systemati-cally fabricated.The mechanical properties(includingflexural strength,elastic modulus and work-of-fracture)were then tested,and analysis of variance(ANOVA)was used for the statistical analysis of the acquired data.The morphol-ogy of FS and the representative fracture surfaces of its filled Bis-GMA/TEGDMA resins/composites were examined by scanning/transmission electron microscopy(SEM/TEM).In addition,the photo-curing behaviors of Bis-GMA/TEGDMA/FS were monitored in situ by using the real time near infrared technique(RT-NIR)to study the photopolymerization kinet-ics(vinyl double bond conversion and photopolymerization rate).2.Materials and methods2.1.MaterialsBis-GMA,TEGDMA,CQ,4EDMAB,MPTMS,n-propylamine and (anhydrous)ethanol were purchased from Sigma–Aldrich Co. (Milwaukee,WI)and used without further purification.Puri-fied FS powder(1250mesh,white/gray in color)was provided by Dalian Global Mineral Co.(Dalian,China).The conventional glassfiller used in this study wasfinely milled7%(mass frac-tion)silanized barium borosilicate glass powder(V-117-2707), provided by Esstech Co.(Essington,PA).d e n t a l m a t e r i a l s 24(2008)235–2432372.2.Fabrication2.2.1.Separation and silanization of FSThe as-received FS (powder)was first dispersed in ethanol with a mass fraction of 5%,and the suspension was then vigorously stirred for 4h at 400rpm using a Heidolph RZR 50Heavy Duty Stirrer.Previous research indicated that this process could effectively separate FS into nano-scaled single crystals,but the separated nano FS single crystals (existing in suspension)could re-aggregate into agglomerates/particles if the ethanol was removed [23].The suspension was then transferred into a rotary evaporator with MPTMS (mass frac-tion of 10%to FS)and n -propylamine (mass fraction of 5%to FS).The system was then heated at 90◦C until dry.It is noted that the silanized FS (powder)might exist primar-ily as agglomerates instead of separated nano-scaled single crystals.2.2.2.Dispersion of silanized FS into dental matrixVarious amounts of silanized FS (mass fractions ranging from 0%to 7.5%)were added into the dental resin sys-tem,which consisted of 49.5%Bis-GMA,49.5%TEGDMA (the mass ratio of Bis-GMA/TEGDMA was 50/50),0.2%CQ and 0.8%4EDMAB.Three dispersion methods were studied to explore the optimal method to distribute the silanized FS ashighly separated nano-scaled single crystals into the dental matrices.Additionally,the prepared suspension systems of Bis-GMA/TEGDMA/FS were further mixed with conventional glass fillers (mass fraction of 50%to Bis-GMA/TEGDMA)to pre-pare the dental pastes.•Method A :The silanized FS was first dispersed into neat TEDGMA and the suspension was vigorously stirred for 2h at 400rpm using the Heidolph RZR 50Heavy Duty Stir-rer.Bis-GMA,CQ and 4EDMAB were then added,and the system was mechanically stirred for another 30min at 400rpm.•Method B :The silanized FS was dispersed into the mix-ture of TEGDMA/ethanol (mass ratio of 10/90),followed by vigorously mechanical stirring for 2h at 400rpm.The ethanol was then removed by vacuum evaporation.Finally,Bis-GMA,CQ and 4EDMAB were added and the system was mechanically stirred for another 30min at 400rpm.•Method C :The silanized FS was dispersed into 90%(mass faction)ethanol diluted dental resin system (composition as described above),followed by vigorously mechanical stir-ring for 2h at 400rpm.The ethanol was then removed by vacuum evaporation and the system was mechanically stirred for another 30min at 400rpm.Fig.1–(a)Representative SEM images of the as-received FS powder.(b)TEM images of separated and silanized FS nano-scaled single crystals.238d e n t a l m a t e r i a l s24(2008)235–2432.3.Characterization and evaluation2.3.1.Photopolymerization kineticsReal time near infrared spectroscopy(RT-NIR)was employed to study the photopolymerization kinetics of Bis-GMA/TEGDMA dental resins containing various mass fractions of the silanized FS.The photopolymerization rate and the degree of vinyl double bond conversion were measured.A Bruker Tensor-27FT-IR spectrometer equipped with a liquid nitrogen-cooled mercury-cadmium-telluride(MCT)detector was employed to carry out the research.During the RT-NIR mea-surements,the spectrometer was continuously purged with dry nitrogen gas,and the uncured resin was placed in a U-shaped Teflon mold with both sides covered by glass slides (note that glass is transparent in the NIR region between the wavenumbers of4000and6500cm−1).The length and width of the mold were4mm,and the thickness was2mm.The glass slides were tightly attached to the mold using small metal clamps.A standard visible light curing unit(Maxima 480),purchased from L.D.Caulk Co.(Milford,DE),was placed directly above the sample holder.The absorption band of vinyl double bond(6100–6250cm−1)were monitored in situ by RT-NIR at ambient condition for5min,with4scans/spectrum and8wavenumber resolution,using series run.In RT-NIR series run,infrared spectra were collected periodically(a1s time interval was used in this study)during the photopoly-merization of the specimens.The curing light was manually turned on at8s and turned off at38s.2.3.2.Mechanical propertiesA standard three-pointflexural test(ASTM D793)with a span of20mm was used to fracture the specimens at a crosshead speed of0.5mm/min using a computer-controlled universal mechanical testing machine(QTEST TM/10,MTS Sys-tems Co.,USA).Analysis of variance(ANOVA),a commercial software(Winks,TexaSoft,Cedar Hill,TX),was used for the statistical analysis of the acquired data.The mechanical test-ing specimens included both Bis-GMA/TEGDMA dental resins (without conventional glassfiller)and composites(with con-ventional glassfiller)containing various mass factions of the silanized FS.Flexural strength,elastic modulus,and work-of-fracture were acquired.Work-of-fracture is the energy required to fracture the specimen and is calculated by dividing the area under the load–displacement curve by the specimen’s cross-sectional area.The dimensions of the Teflon molds for making the three-pointflexural testing specimens were 2mm×2mm×25mm.The specimens were photo-cured for 1min on each side,and then carefully removed from the molds.Prior to mechanical testing,the specimens were stored in a humidifier at37◦C for24h.Six specimens were prepared for each measurement,and all four sides of each specimen were carefully hand-polished with2400and4000grit silicon carbide paper and water coolant in a longitudinal direction.2.3.3.MorphologyA Zeiss Supra40VPfield-emission scanning electron micro-scope(SEM)and a Hitachi H-7000FA transmission electron microscope(TEM)were employed to examine the morpholo-gies of FS and the representative fracture surfaces of the Bis-GMA/TEGDMA dental resins/composites containing var-ious mass fractions of nano FS.Prior to SEM examination, the specimens were sputter-coated with gold to avoid charge accumulation.For TEM examination,the specimen was micro-tomed at room temperature using a Reichert-Jung Ultracut Microtome and mounted on200mesh copper grids.3.Results and discussion3.1.Separation and silanization of FSNaturally occurring FS minerals usually contain some impu-rities including silica and carbonates.These impuritieswereFig.2–SEM images of representative fracture surfaces of 2.5%(mass fraction)nano FSfilled Bis-GMA/TEGDMA dental resins.The letters A,B and C represent different dispersion methods.d e n t a l m a t e r i a l s24(2008)235–243239removed in the as-received FS(powder).Fig.1a shows the rep-resentative SEM images of the as-received FS.It was evident that the FS agglomerates/particles consisted of nano-scaled single crystals(fibers)with diameters in tens of nanometers and lengths in microns.Some of the FS nano-scaled single crystals stacked together into bundles,and thefibers/bundles then aggregated into agglomerates.The FS agglomerates ranged from submicron to several microns in size.The as-received FS(powder)was silanized using the proce-dure as described in the experimental section.Fig.1b shows the TEM images of the(silanized and separated)FS nano-scaled single crystals(fibers).To prepare the TEM samples, the silanized FS wasfirst dispersed in ethanol with a mass fraction of approximately1%,and the suspension was then mechanically stirred for30min at400rpm using the Heidolph RZR50Heavy Duty Stirrer.Subsequently,the carbon-coated TEM grids were dipped into the uniform suspension(with no clearly identifiable solid/precipitate)and quickly removed. After the ethanol in the suspension left on the TEM grids evaporated,the samples were used for TEM examination.It is noted that there might be agglomerates in the silanized FS(powder)although the silanized FS was well separated as nano-scaled single crystals in Fig.1b(presumably due to the rapid evaporation rate of ethanol in the method to prepare the TEM samples).3.2.Dispersion of silanized FS into dental matrixTo uniformly disperse nano-scaledfillers into dental matri-ces is always a pared to layered silicates such as montmorillonite,FS is considerably easier to separate into nano-scaled single crystals(fibers),as evidenced in Fig.1b. Nonetheless,it was still a challenge to achieve the high degree of separation and uniform dispersion of the silanized FS in Bis-GMA/TEGDMA.This is because the viscosity of Bis-GMA/TEGDMA is orders of magnitude higher than that of ethanol and the polarity is much lower.Thus,although the silanized FS could be highly separated and uniformly dis-persed in ethanol,it might not be able to achieve the same high degree of separation and dispersion in Bis-GMA/TEGDMA.To explore the optimal method/procedure for dispers-ing the silanized FS in Bis-GMA/TEGDMA,three methods (as detailed in the Section2)were studied.Fig.2shows the representative fracture surfaces of the photo-cured Bis-GMA/TEGDMA(50/50mass ratio)dental resinsfilled with2.5% (mass fraction)silanized FS.The letters A,B and C represent the three different dispersion methods.The specimens used for Fig.2were also microtomed for TEM examination,and the results are shown in Fig.3.As shown in Figs.2A and3A,most of the silanized FS in the Bis-GMA/TEGDMA dental resin prepared by“Method A”existed as agglomerates,and only trace amount of FS seemed to exist as separated single crystals(indicated by tiny bright dots in Fig.2A,with almost no separated single crystal identified in the TEM image).This suggested that direct dispersion of the silanized FS in dental monomers (even in the diluent monomer of TEGDMA alone)could barely separate FS into nano-scaled single crystals.In comparison,“Method B”and“C”resulted in a much higher degree of separation and dispersion,since many FS nano-scaled single crystals were clearly identifiable in the SEM/TEM images.The results suggested that the use of ethanol during the disper-sion process could greatly improve the degree of separation and dispersion of the silanized FS.Nonetheless,there were still agglomerates found in Fig.3B and C.This suggested that,unlike pure ethanol,the ethanol solutions containing TEGDMA and/or Bis-GMA/TEGDMA might not be able to well separate the silanized FS or,after the removal/evaporation of ethanol,some of the separated FS nano-scaled single crystals might re-aggregate into agglomerates.Further investigations (including studies on different surface treatment/silanization techniques)are being conducted to identify better dispersing methods/procedures.Compared to“Method B”,“Method C”seemed more effective since more separated FS nano-scaled single crystals (rather than agglomerates)were present in the SEM/TEM images(see Fig.2B and C and Fig.3B and C).We also attempted to further improve the separation and dispersion of the silanized FS in the Bis-GMA/TEGDMA/FS system(pre-pared by“Method C”)by using an ultrasonic probe purchased from the Fisher Scientific Co.(model number500),but no distinguishable difference was identified(results not shown).“Method C”was thus adopted as the dispersion method to prepare specimens for the photopolymerizationkineticsFig.3–TEM images of2.5%(mass fraction)nano FSfilled Bis-GMA/TEGDMA dental resins.The letters A,B and C represent different dispersion methods.240d e n t a l m a t e r i a l s24(2008)235–243Fig.4–(a)RT-NIR spectra of2.5%(mass fraction)nano FSfilled Bis-GMA/TEGDMA resin after photo-curing for30s. Spectra were collected at different times,from top to bottom,8s(immediately before the curing light was turned on),12s,16s,60s,respectively.(b)RT-NIR photo-curingprofiles of Bis-GMA/TEGDMA dental resinsfilled with various mass fractions of nano FS.The profiles are offset for clarity.studies and the mechanical property measurements.We note that the agglomerates with loosely-stacked FS nano-scaled single crystals could act as the mechanical weak points(struc-tural defects);and the presence of such agglomerates could significantly reduce the mechanical properties(especially strength)of the nano FS reinforced composites.3.3.Photopolymerization kineticsPhotopolymerization kinetics was studied to determine if Bis-GMA/TEGDMA dental resins containing nano FS could be appropriately photo-cured.The real time near infrared(RT-NIR)series run technique(as detailed in the Section2)was adopted for this study.RT-NIR spectroscopic technique mea-sured the conversion of vinyl double bond in dental resins by following the overtone band located from6100to6250cm−1. The use of this technique to monitor the photo-curing behav-iors of dental methacrylate-based resins was demonstrated by Stansbury and Dickens[24].The RT-NIR series run provided valuable insights into the photopolymerization rate as well as the vinyl double bond conversion.Fig.4a shows the RT-NIR spectra of2.5%(mass fraction) nano FSfilled Bis-GMA/TEGDMA dental resin after the photo-curing for30s.From top to bottom,spectra were collected at8s (immediately before turning on the curing light),and12s,16s, 60s,respectively.The spectra indicated that the vinyl dou-ble bond absorbance varied in situ with photopolymerization. After the curing light was turned on(i.e.,the photopolymer-ization was initiated),the absorbance dropped quickly during the initial10s.The absorbance remained at approximately the same level for the rest of the period.The absorbance peak in Fig.4a was integrated to obtain the NIR curing profiles.By using the absorbance of the vinyl double bond beforepho-Fig.5–Mechanical properties:(a)flexural strength;(b) elastic modulus;and(c)work-of-fracture,ofBis-GMA/TEGDMA dental resins/compositesfilled with various mass fractions of nano FS.Each datum is the mean value of six measurements with error bar representing one standard deviation.d e n t a l m a t e r i a l s24(2008)235–243241topolymerization as the reference,the NIR curing profiles were converted to the degrees of vinyl double bond conversion,as depicted in Fig.4b.The tangent of the curves in Fig.4b at each time spot represents the photopolymerization rate at that par-ticular time.Fig.4a and b indicated that all the systems of Bis-GMA/TEGDMA dental resinsfilled with various mass frac-tions(up to7.5%)of nano FS could be photo-cured in less than 15s to achieve an approximately90%vinyl double bond con-version.The addition of nano FS into Bis-GMA/TEGDMA did not significantly vary either the vinyl double bond conversion or the photopolymerization rate.3.4.Mechanical propertiesFlexural strength(S F),elastics modulus(E Y)and work-of-fracture(WOF)of Bis-GMA/TEGDMA(50/50mass ratio)dental resins/composites containing various mass fractions of the silanized nano FS were tested,and the results are showed in Fig.5.The control samples were the resins/composites with-out the nano FS.Each datum in the plots provides the mean value of six measurements with the error bar representing one standard deviation.As shown in Fig.5,the values of S F,E Y and WOF were all substantially increased by the impregnation of small mass fractions of the nano FS into the Bis-GMA/TEGDMA dental resins(without conventional glassfiller).S F,E Y and WOF for the unfilled/neat resin(mean±standard deviation,n=6)were (90±4)MPa,(1.8±0.2)GPa and(5.1±0.8)kJ/m2,respectively. For the resinfilled with1.0%(mass fraction)nano FS,S F,E Y and WOF were increased to(126±4)MPa,(2.1±0.2)GPa and (9.1±1.0)kJ/m2,respectively.Thus,theflexural strength was improved by40%,the elastic modulus was improved by16.7%, and the work-of-fracture was improved by78.4%.However, increasing the mass fractions of the nano FS did not fur-ther improve the mechanical properties.For the resinfilled with2.5%(mass fraction)nano FS,the measured S F,E Y and WOF values were,respectively,(128±6)MPa,(2.4±0.4)GPa and(8.6±1.1)kJ/m2.For the resinfilled with7.5%(mass frac-tion)nano FS,the respective S F E Y and WOF values were (133±7)MPa,(2.8±0.3)GPa and(7.9±1.1)kJ/m2.These results indicated that the impregnation of the silanized nano FS into Bis-GMA/TEGDMA dental resin could significantly improve the mechanical properties.However,the nano FS has to be highly separated into nano-scaled single crystals and uniformly dis-tributed in the dental resin.An effective reinforcement would not be achieved if the nano FS existed primarily as agglom-erates/particles which could actually act as the mechanical weak points(structural defects).Representative fracture surfaces of the neat and nano FS reinforced Bis-GMA/TEGDMA dental resins are shown in Fig.6a and b.The neat resin(Fig.6a)fractured as a typi-cal ductile resin,and the fracture surface was smooth with oriented fracture lines resulted from the extension of craz-ings initiated by the stress concentration points.On the other hand,the fracture surface of the nano FS reinforced resin (Fig.6b)was rough with no clearly identifiable fracture lines. The results suggested that the presence of nano FS could deflect the micro-crack and effectively increase the resis-tance to the applied force.When the crackfinally broke away from the nano FS,a rough fracture surface was created,sug-gesting energy consumption during breaking.However,some voids/holes were also observed on the fracture surface oftheFig.6–Representative fracture surfaces of three-pointflexural specimens:(a)neat/unfilled Bis-GMA/TEGDMA;(b)Bis-GMA/TEGDMAfilled with2.5%(mass faction)nano FS;(c)Bis-GMA/TEGDMAfilled with50%(mass faction)glassfiller;(d) Bis-GMA/TEGDMAfilled with2.5%(mass faction)nano FS and50%(mass faction)glassfiller.242d e n t a l m a t e r i a l s24(2008)235–243nano FS reinforced resin.These voids/holes were formed by the failure/fallout of the FS agglomerates/particles.As dis-cussed above,the agglomerates could act as structural defects and significantly weaken thefilled resin.Taken together,the impregnation of the silanized nano FS into Bis-GMA/TEGDMA dental resin may result in two opposite effects:the reinforcing effect due to the highly separated and well distributed nano FS single crystals,and the weakening effect due to the formation of FS agglomerates/particles.If the nano FS could be uni-formly distributed in the dental matrices as highly separated nano-scaled single crystals,we envision that the mechanical properties of the resulting composites would be significantly higher than those of the currently developed composites.The mechanical properties of the Bis-GMA/TEGDMA den-tal composites(with conventional glassfiller)containing small mass fractions of the silanized nano FS showed a similar trend of improvement.Bothflexural strengths and elastics moduli of the compositesfilled with1%and2.5%(mass fractions)of the nano FS were higher than those of the control sample. Nonetheless,the compositesfilled with7.5%(mass faction)of the silanized nano FS had S F and E Y values of(101±6)MPa and(3.7±0.4)GPa,which were statistically the same as those of the control sample(one way ANOVA,P>0.05).For the work-of-fracture values,one way ANOVA indicated no statis-tical difference between the control sample and the samples filled with1%,2.5%and7.5%(mass factions)nano FS.These results further supported the previous conclusion that the impregnation of the silanized nano FS into Bis-GMA/TEGDMA dental matrices could result in both reinforcing and weak-ening effects,depending upon the degree of separation and uniformity of distribution of the nano FS.Unlike that in resins, the nano FS seemed more likely to aggregate into agglom-erates and the sizes of the agglomerates seemed larger in composites,as indicated by the SEM images of the represen-tative fracture surfaces(Fig.6).The detailed reasons for this difference are still under investigation.4.ConclusionsThe aim of this study was to investigate the reinforcement of Bis-GMA/TEGDMA dental resins/composites with various mass fractions of nano FS.FS is a natural mineral com-posed of nano-scaled silicate single crystals(fibers)with diameters in tens of nanometers and lengths in microns. Impregnation of small mass fractions(1%and2.5%)of the silanized nano FS into Bis-GMA/TEGDMA(50/50mass ratio) dental resins/composites improved the mechanical proper-ties rger mass fraction of impregnation(7.5%), however,did not further improve the mechanical properties (one way ANOVA,P>0.05),and may even have reduced the mechanical properties.Impregnation of the nano FS into Bis-GMA/TEGDMA dental resins/composites could result in two opposite effects:the reinforcing effect due to the highly sep-arated and uniformly distributed nano FS single crystals,and the weakening effect due to the formation of FS agglomer-ates/particles.The simultaneous accomplishment of a high degree separation and uniform dispersion of nano FS in dental resins/composites is still under investigations.Nonetheless, since the impregnation of small mass fractions of the nano FS into the dental resins/composites could effectively improve the mechanical properties,nano FS may have significant value to be used as the reinforcing nanofiller for dental composites.AcknowledgementThis research was supported by the National Institute of Dental and Craniofacial Research(R03DE16042),and by the “Center for Accelerated Applications at the Nanoscale(CAAN)”and the“BioMedical Engineering(BME)Program”at the South Dakota School of Mines and Technology(SDSM&T).The authors are also grateful to Esstech Co.for providing the silanized barium borosilicate glassfiller.r e f e r e n c e s[1]Bowen RL.,Dentalfilling material comprising vinyl-silanetreated fused silica and a binder consisting of the reactionproduct of bisphenol and glycidyl methacrylate.U.S.Pa.3,066,112.1962.[2]Bowen RL.Properties of a silica-reinforced polymer fordental restoration.J Am Dent Assoc1963;66:57–64.[3]Antonucci JM,Stansbury JW.Molecular designed dentalpolymer.In:Arshady R,editor.Desk reference of functional polymers:synthesis and application.American ChemicalSociety Publication;1997.p.719–38.[4]Reed BB,Choi K,Dickens SH,Stansbury JW.Effect of resincomposition of kinetics of dimethacrylatephotopolymerization.Polym Prepr1997;38(2):108–9.[5]Leinfelder KF,Sluder TB,Santos JFF,Wall JT.Five-yearclinical evaluation of anterior and posterior restorations ofcomposite resin.Oper Dent1980;12:52–78.[6]Lacy AM.A critical look at posterior composite restorations.J Am Dent Assoc1987;114:357–62.[7]Jordan RE,Suzuki M.Posterior composite restorations.J AmDent Assoc1991;122:31–7.[8]O’Brien WJ.Dental materials and their selection.3rd ed.Quintessence Publishing Co Inc.;2002.[9]Corbin SB,Kohn WG.The benefits and risks of dentalamalgam.J Am Dent Assoc1994;125:381–8.[10]Berry TG,Nicholson J,Troendle K.Almost two centuries withamalgam,where are we today?J Am Dent Assoc1994;125:392–9.[11]Kusy RP,Leinfelder KF.Pattern of wear in posteriorcomposite restorations.J Dent Res1977;56(5):544–1544. [12]Abell AK,Leinfelder KF,T urner DT.Microscopic observationsof the wear of a tooth restorative composite in vivo.JBiomed Mat Res1983;17(3):501–7.[13]Grave AMH,Chandler HD,Wolfaardt JF.Denture base acrylicreinforced with high modulusfibre.Dent Mater1985;1:185–7.[14]Goldberg AJ,Burstone CJ.The use of continuousfiberreinforcement in dentistry.Dent Mater1992;8:197–202. [15]Giordano R.Fiber reinforced composite resin systems.GenDent2000;48(3):244–9.[16]Behr parison of three types offiber-reinforcedcomposite molar crowns on their fracture resistance andmarginal adaptation.J Dent2001;29(3):187–96.[17]Xu HHK,Martin TA,Antonucci JM,Eichmiller FC.Ceramicwhisker reinforcement of dental composite resins.J DentRes1999;78(2):706–12.[18]Xu HHK.Dental composite resins containing silica-fusedceramic single-crystalline whiskers with variousfiller levels.J Dent Res1999;78:1304–11.。