Models for Low-Mass X-Ray Binaries in the Elliptical Galaxies NGC3379 and NGC4278 Compariso

合集下载

Autodesk Nastran 2022 用户手册说明书

DATINFILE2 ......................................................................................................................................................... 10
MPA, MPI (design/logo), MPX (design/logo), MPX, Mudbox, Navisworks, ObjectARX, ObjectDBX, Opticore, Pixlr, Pixlr-o-matic, Productstream,
Publisher 360, RasterDWG, RealDWG, ReCap, ReCap 360, Remote, Revit LT, Revit, RiverCAD, Robot, Scaleform, Showcase, Showcase 360,
TrueConvert, DWG TrueView, DWGX, DXF, Ecotect, Ember, ESTmep, Evolver, FABmep, Face Robot, FBX, Fempro, Fire, Flame, Flare, Flint,
ForceEffect, FormIt, Freewheel, Fusion 360, Glue, Green Building Studio, Heidi, Homestyler, HumanIK, i-drop, ImageModeler, Incinerator, Inferno,
Autodesk Nastran 2022
Reference Manual
Nastran Solver Reference Manual

Conditional Random Fields_ Probabilistic Models for Segmenting and Labeling Sequence Data

Conditional Random Fields:Probabilistic Modelsfor Segmenting and Labeling Sequence DataJohn Lafferty LAFFERTY@ Andrew McCallum MCCALLUM@ Fernando Pereira FPEREIRA@ WhizBang!Labs–Research,4616Henry Street,Pittsburgh,PA15213USASchool of Computer Science,Carnegie Mellon University,Pittsburgh,PA15213USADepartment of Computer and Information Science,University of Pennsylvania,Philadelphia,PA19104USAAbstractWe present,a frame-work for building probabilistic models to seg-ment and label sequence data.Conditional ran-domﬁelds offer several advantages over hid-den Markov models and stochastic grammarsfor such tasks,including the ability to relaxstrong independence assumptions made in thosemodels.Conditional randomﬁelds also avoida fundamental limitation of maximum entropyMarkov models(MEMMs)and other discrimi-native Markov models based on directed graph-ical models,which can be biased towards stateswith few successor states.We present iterativeparameter estimation algorithms for conditionalrandomﬁelds and compare the performance ofthe resulting models to HMMs and MEMMs onsynthetic and natural-language data.1.IntroductionThe need to segment and label sequences arises in many different problems in several scientiﬁcﬁelds.Hidden Markov models(HMMs)and stochastic grammars are well understood and widely used probabilistic models for such problems.In computational biology,HMMs and stochas-tic grammars have been successfully used to align bio-logical sequences,ﬁnd sequences homologous to a known evolutionary family,and analyze RNA secondary structure (Durbin et al.,1998).In computational linguistics and computer science,HMMs and stochastic grammars have been applied to a wide variety of problems in text and speech processing,including topic segmentation,part-of-speech(POS)tagging,information extraction,and syntac-tic disambiguation(Manning&Sch¨u tze,1999).HMMs and stochastic grammars are generative models,as-signing a joint probability to paired observation and label sequences;the parameters are typically trained to maxi-mize the joint likelihood of training examples.To deﬁne a joint probability over observation and label sequences, a generative model needs to enumerate all possible ob-servation sequences,typically requiring a representation in which observations are task-appropriate atomic entities, such as words or nucleotides.In particular,it is not practi-cal to represent multiple interacting features or long-range dependencies of the observations,since the inference prob-lem for such models is intractable.This difﬁculty is one of the main motivations for looking at conditional models as an alternative.A conditional model speciﬁes the probabilities of possible label sequences given an observation sequence.Therefore,it does not expend modeling effort on the observations,which at test time areﬁxed anyway.Furthermore,the conditional probabil-ity of the label sequence can depend on arbitrary,non-independent features of the observation sequence without forcing the model to account for the distribution of those dependencies.The chosen features may represent attributes at different levels of granularity of the same observations (for example,words and characters in English text),or aggregate properties of the observation sequence(for in-stance,text layout).The probability of a transition between labels may depend not only on the current observation, but also on past and future observations,if available.In contrast,generative models must make very strict indepen-dence assumptions on the observations,for instance condi-tional independence given the labels,to achieve tractability. Maximum entropy Markov models(MEMMs)are condi-tional probabilistic sequence models that attain all of the above advantages(McCallum et al.,2000).In MEMMs, each source state1has a exponential model that takes the observation features as input,and outputs a distribution over possible next states.These exponential models are trained by an appropriate iterative scaling method in the 1Output labels are associated with states;it is possible for sev-eral states to have the same label,but for simplicity in the rest of this paper we assume a one-to-one correspondence.maximum entropy framework.Previously published exper-imental results show MEMMs increasing recall and dou-bling precision relative to HMMs in a FAQ segmentation task.MEMMs and other non-generativeﬁnite-state models based on next-state classiﬁers,such as discriminative Markov models(Bottou,1991),share a weakness we callhere the:the transitions leaving a given state compete only against each other,rather than againstall other transitions in the model.In probabilistic terms, transition scores are the conditional probabilities of pos-sible next states given the current state and the observa-tion sequence.This per-state normalization of transition scores implies a“conservation of score mass”(Bottou, 1991)whereby all the mass that arrives at a state must be distributed among the possible successor states.An obser-vation can affect which destination states get the mass,but not how much total mass to pass on.This causes a bias to-ward states with fewer outgoing transitions.In the extreme case,a state with a single outgoing transition effectively ignores the observation.In those cases,unlike in HMMs, Viterbi decoding cannot downgrade a branch based on ob-servations after the branch point,and models with state-transition structures that have sparsely connected chains of states are not properly handled.The Markovian assump-tions in MEMMs and similar state-conditional models in-sulate decisions at one state from future decisions in a way that does not match the actual dependencies between con-secutive states.This paper introduces(CRFs),a sequence modeling framework that has all the advantages of MEMMs but also solves the label bias problem in a principled way.The critical difference between CRFs and MEMMs is that a MEMM uses per-state exponential mod-els for the conditional probabilities of next states given the current state,while a CRF has a single exponential model for the joint probability of the entire sequence of labels given the observation sequence.Therefore,the weights of different features at different states can be traded off against each other.We can also think of a CRF as aﬁnite state model with un-normalized transition probabilities.However,unlike some other weightedﬁnite-state approaches(LeCun et al.,1998), CRFs assign a well-deﬁned probability distribution over possible labelings,trained by maximum likelihood or MAP estimation.Furthermore,the loss function is convex,2guar-anteeing convergence to the global optimum.CRFs also generalize easily to analogues of stochastic context-free grammars that would be useful in such problems as RNA secondary structure prediction and natural language pro-cessing.2In the case of fully observable states,as we are discussing here;if several states have the same label,the usual local maxima of Baum-Welch arise.bel bias example,after(Bottou,1991).For concise-ness,we place observation-label pairs on transitions rather than states;the symbol‘’represents the null output label.We present the model,describe two training procedures and sketch a proof of convergence.We also give experimental results on synthetic data showing that CRFs solve the clas-sical version of the label bias problem,and,more signiﬁ-cantly,that CRFs perform better than HMMs and MEMMs when the true data distribution has higher-order dependen-cies than the model,as is often the case in practice.Finally, we conﬁrm these results as well as the claimed advantages of conditional models by evaluating HMMs,MEMMs and CRFs with identical state structure on a part-of-speech tag-ging task.2.The Label Bias ProblemClassical probabilistic automata(Paz,1971),discrimina-tive Markov models(Bottou,1991),maximum entropy taggers(Ratnaparkhi,1996),and MEMMs,as well as non-probabilistic sequence tagging and segmentation mod-els with independently trained next-state classiﬁers(Pun-yakanok&Roth,2001)are all potential victims of the label bias problem.For example,Figure1represents a simpleﬁnite-state model designed to distinguish between the two words and.Suppose that the observation sequence is. In theﬁrst time step,matches both transitions from the start state,so the probability mass gets distributed roughly equally among those two transitions.Next we observe. Both states1and4have only one outgoing transition.State 1has seen this observation often in training,state4has al-most never seen this observation;but like state1,state4 has no choice but to pass all its mass to its single outgoing transition,since it is not generating the observation,only conditioning on it.Thus,states with a single outgoing tran-sition effectively ignore their observations.More generally, states with low-entropy next state distributions will take lit-tle notice of observations.Returning to the example,the top path and the bottom path will be about equally likely, independently of the observation sequence.If one of the two words is slightly more common in the training set,the transitions out of the start state will slightly prefer its cor-responding transition,and that word’s state sequence will always win.This behavior is demonstrated experimentally in Section5.L´e on Bottou(1991)discussed two solutions for the label bias problem.One is to change the state-transition struc-ture of the model.In the above example we could collapse states1and4,and delay the branching until we get a dis-criminating observation.This operation is a special case of determinization(Mohri,1997),but determinization of weightedﬁnite-state machines is not always possible,and even when possible,it may lead to combinatorial explo-sion.The other solution mentioned is to start with a fully-connected model and let the training procedureﬁgure out a good structure.But that would preclude the use of prior structural knowledge that has proven so valuable in infor-mation extraction tasks(Freitag&McCallum,2000). Proper solutions require models that account for whole state sequences at once by letting some transitions“vote”more strongly than others depending on the corresponding observations.This implies that score mass will not be con-served,but instead individual transitions can“amplify”or “dampen”the mass they receive.In the above example,the transitions from the start state would have a very weak ef-fect on path score,while the transitions from states1and4 would have much stronger effects,amplifying or damping depending on the actual observation,and a proportionally higher contribution to the selection of the Viterbi path.3In the related work section we discuss other heuristic model classes that account for state sequences globally rather than locally.To the best of our knowledge,CRFs are the only model class that does this in a purely probabilistic setting, with guaranteed global maximum likelihood convergence.3.Conditional Random FieldsIn what follows,is a random variable over data se-quences to be labeled,and is a random variable over corresponding label sequences.All components of are assumed to range over aﬁnite label alphabet.For ex-ample,might range over natural language sentences and range over part-of-speech taggings of those sentences, with the set of possible part-of-speech tags.The ran-dom variables and are jointly distributed,but in a dis-criminative framework we construct a conditional model from paired observation and label sequences,and do not explicitly model the marginal..Thus,a CRF is a randomﬁeld globally conditioned on the observation.Throughout the paper we tacitly assume that the graph isﬁxed.In the simplest and most impor-3Weighted determinization and minimization techniques shift transition weights while preserving overall path weight(Mohri, 2000);their connection to this discussion deserves further study.tant example for modeling sequences,is a simple chain or line:.may also have a natural graph structure;yet in gen-eral it is not necessary to assume that and have the same graphical structure,or even that has any graph-ical structure at all.However,in this paper we will be most concerned with sequencesand.If the graph of is a tree(of which a chain is the simplest example),its cliques are the edges and ver-tices.Therefore,by the fundamental theorem of random ﬁelds(Hammersley&Clifford,1971),the joint distribu-tion over the label sequence given has the form(1),where is a data sequence,a label sequence,and is the set of components of associated with the vertices in subgraph.We assume that the and are given andﬁxed. For example,a Boolean vertex feature might be true if the word is upper case and the tag is“proper noun.”The parameter estimation problem is to determine the pa-rameters from training datawith empirical distribution. In Section4we describe an iterative scaling algorithm that maximizes the log-likelihood objective function:.As a particular case,we can construct an HMM-like CRF by deﬁning one feature for each state pair,and one feature for each state-observation pair:. The corresponding parameters and play a simi-lar role to the(logarithms of the)usual HMM parameters and.Boltzmann chain models(Saul&Jor-dan,1996;MacKay,1996)have a similar form but use a single normalization constant to yield a joint distribution, whereas CRFs use the observation-dependent normaliza-tion for conditional distributions.Although it encompasses HMM-like models,the class of conditional randomﬁelds is much more expressive,be-cause it allows arbitrary dependencies on the observationFigure2.Graphical structures of simple HMMs(left),MEMMs(center),and the chain-structured case of CRFs(right)for sequences. An open circle indicates that the variable is not generated by the model.sequence.In addition,the features do not need to specify completely a state or observation,so one might expect that the model can be estimated from less training data.Another attractive property is the convexity of the loss function;in-deed,CRFs share all of the convexity properties of general maximum entropy models.For the remainder of the paper we assume that the depen-dencies of,conditioned on,form a chain.To sim-plify some expressions,we add special start and stop states and.Thus,we will be using the graphical structure shown in Figure2.For a chain struc-ture,the conditional probability of a label sequence can be expressed concisely in matrix form,which will be useful in describing the parameter estimation and inference al-gorithms in Section4.Suppose that is a CRF given by(1).For each position in the observation se-quence,we deﬁne the matrix random variableby,where is the edge with labels and is the vertex with label.In contrast to generative models,con-ditional models like CRFs do not need to enumerate over all possible observation sequences,and therefore these matrices can be computed directly as needed from a given training or test observation sequence and the parameter vector.Then the normalization(partition function)is the entry of the product of these matrices:. Using this notation,the conditional probability of a label sequence is written as, where and.4.Parameter Estimation for CRFsWe now describe two iterative scaling algorithms toﬁnd the parameter vector that maximizes the log-likelihood of the training data.Both algorithms are based on the im-proved iterative scaling(IIS)algorithm of Della Pietra et al. (1997);the proof technique based on auxiliary functions can be extended to show convergence of the algorithms for CRFs.Iterative scaling algorithms update the weights asand for appropriately chosen and.In particular,the IIS update for an edge feature is the solution ofdef.where is thedef. The equations for vertex feature updates have similar form.However,efﬁciently computing the exponential sums on the right-hand sides of these equations is problematic,be-cause is a global property of,and dynamic programming will sum over sequences with potentially varying.To deal with this,theﬁrst algorithm,Algorithm S,uses a“slack feature.”The second,Algorithm T,keepstrack of partial totals.For Algorithm S,we deﬁne the bydef,where is a constant chosen so that for all and all observation vectors in the training set,thus making.Feature is“global,”that is,it does not correspond to any particular edge or vertex.For each index we now deﬁne thewith base caseifotherwiseand recurrence.Similarly,the are deﬁned byifotherwiseand.With these deﬁnitions,the update equations are,where.The factors involving the forward and backward vectors in the above equations have the same meaning as for standard hidden Markov models.For example,is the marginal probability of label given that the observation sequence is.This algorithm is closely related to the algorithm of Darroch and Ratcliff(1972),and MART algorithms used in image reconstruction.The constant in Algorithm S can be quite large,since in practice it is proportional to the length of the longest train-ing observation sequence.As a result,the algorithm may converge slowly,taking very small steps toward the maxi-mum in each iteration.If the length of the observations and the number of active features varies greatly,a faster-converging algorithm can be obtained by keeping track of feature totals for each observation sequence separately. Let def.Algorithm T accumulates feature expectations into counters indexed by.More speciﬁcally,we use the forward-backward recurrences just introduced to compute the expectations of feature and of feature given that.Then our param-eter updates are and,whereand are the unique positive roots to the following polynomial equationsmax max,(2)which can be easily computed by Newton’s method.A single iteration of Algorithm S and Algorithm T has roughly the same time and space complexity as the well known Baum-Welch algorithm for HMMs.To prove con-vergence of our algorithms,we can derive an auxiliary function to bound the change in likelihood from below;this method is developed in detail by Della Pietra et al.(1997). The full proof is somewhat detailed;however,here we give an idea of how to derive the auxiliary function.To simplify notation,we assume only edge features with parameters .Given two parameter settings and,we bound from below the change in the objective function with anas followsdefwhere the inequalities follow from the convexity of and.Differentiating with respect to and setting the result to zero yields equation(2).5.ExperimentsWeﬁrst discuss two sets of experiments with synthetic data that highlight the differences between CRFs and MEMMs. Theﬁrst experiments are a direct veriﬁcation of the label bias problem discussed in Section2.In the second set of experiments,we generate synthetic data using randomly chosen hidden Markov models,each of which is a mix-ture of aﬁrst-order and second-order peting models are then trained and compared on test data.As the data becomes more second-order,the test er-ror rates of the trained models increase.This experiment corresponds to the common modeling practice of approxi-mating complex local and long-range dependencies,as oc-cur in natural data,by small-order Markov models.OurFigure3.Plots of error rates for HMMs,CRFs,and MEMMs on randomly generated synthetic data sets,as described in Section5.2. As the data becomes“more second order,”the error rates of the test models increase.As shown in the left plot,the CRF typically signiﬁcantly outperforms the MEMM.The center plot shows that the HMM outperforms the MEMM.In the right plot,each open square represents a data set with,and a solid circle indicates a data set with.The plot shows that when the data is mostly second order(),the discriminatively trained CRF typically outperforms the HMM.These experiments are not designed to demonstrate the advantages of the additional representational power of CRFs and MEMMs relative to HMMs.results clearly indicate that even when the models are pa-rameterized in exactly the same way,CRFs are more ro-bust to inaccurate modeling assumptions than MEMMs or HMMs,and resolve the label bias problem,which affects the performance of MEMMs.To avoid confusion of dif-ferent effects,the MEMMs and CRFs in these experiments use overlapping features of the observations.Fi-nally,in a set of POS tagging experiments,we conﬁrm the advantage of CRFs over MEMMs.We also show that the addition of overlapping features to CRFs and MEMMs al-lows them to perform much better than HMMs,as already shown for MEMMs by McCallum et al.(2000).5.1Modeling label biasWe generate data from a simple HMM which encodes a noisy version of theﬁnite-state network in Figure1.Each state emits its designated symbol with probabilityand any of the other symbols with probability.We train both an MEMM and a CRF with the same topologies on the data generated by the HMM.The observation fea-tures are simply the identity of the observation symbols. In a typical run using training and test samples, trained to convergence of the iterative scaling algorithm, the CRF error is while the MEMM error is, showing that the MEMM fails to discriminate between the two branches.5.2Modeling mixed-order sourcesFor these results,we useﬁve labels,(),and26 observation values,();however,the results were qualitatively the same over a range of sizes for and .We generate data from a mixed-order HMM with state transition probabilities given byand,simi-larly,emission probabilities given by.Thus,for we have a standardﬁrst-order HMM.In order to limit the size of the Bayes error rate for the resulting models,the con-ditional probability tables are constrained to be sparse. In particular,can have at most two nonzero en-tries,for each,and can have at most three nonzero entries for each.For each randomly gener-ated model,a sample of1,000sequences of length25is generated for training and testing.On each randomly generated training set,a CRF is trained using Algorithm S.(Note that since the length of the se-quences and number of active features is constant,Algo-rithms S and T are identical.)The algorithm is fairly slow to converge,typically taking approximately500iterations for the model to stabilize.On the500MHz Pentium PC used in our experiments,each iteration takes approximately 0.2seconds.On the same data an MEMM is trained using iterative scaling,which does not require forward-backward calculations,and is thus more efﬁcient.The MEMM train-ing converges more quickly,stabilizing after approximately 100iterations.For each model,the Viterbi algorithm is used to label a test set;the experimental results do not sig-niﬁcantly change when using forward-backward decoding to minimize the per-symbol error rate.The results of several runs are presented in Figure3.Each plot compares two classes of models,with each point indi-cating the error rate for a single test set.As increases,the error rates generally increase,as theﬁrst-order models fail toﬁt the second-order data.Theﬁgure compares models parameterized as,,and;results for models parameterized as,,and are qualitatively the same.As shown in theﬁrst graph,the CRF generally out-performs the MEMM,often by a wide margin of10%–20% relative error.(The points for very small error rate,with ,where the MEMM does better than the CRF, are suspected to be the result of an insufﬁcient number of training iterations for the CRF.)HMM 5.69%45.99%MEMM 6.37%54.61%CRF 5.55%48.05%MEMM 4.81%26.99%CRF 4.27%23.76%Using spelling featuresFigure4.Per-word error rates for POS tagging on the Penn tree-bank,usingﬁrst-order models trained on50%of the1.1million word corpus.The oov rate is5.45%.5.3POS tagging experimentsTo conﬁrm our synthetic data results,we also compared HMMs,MEMMs and CRFs on Penn treebank POS tag-ging,where each word in a given input sentence must be labeled with one of45syntactic tags.We carried out two sets of experiments with this natural language data.First,we trainedﬁrst-order HMM,MEMM, and CRF models as in the synthetic data experiments,in-troducing parameters for each tag-word pair andfor each tag-tag pair in the training set.The results are con-sistent with what is observed on synthetic data:the HMM outperforms the MEMM,as a consequence of the label bias problem,while the CRF outperforms the HMM.The er-ror rates for training runs using a50%-50%train-test split are shown in Figure5.3;the results are qualitatively sim-ilar for other splits of the data.The error rates on out-of-vocabulary(oov)words,which are not observed in the training set,are reported separately.In the second set of experiments,we take advantage of the power of conditional models by adding a small set of or-thographic features:whether a spelling begins with a num-ber or upper case letter,whether it contains a hyphen,and whether it ends in one of the following sufﬁxes:.Here weﬁnd,as expected,that both the MEMM and the CRF beneﬁt signif-icantly from the use of these features,with the overall error rate reduced by around25%,and the out-of-vocabulary er-ror rate reduced by around50%.One usually starts training from the all zero parameter vec-tor,corresponding to the uniform distribution.However, for these datasets,CRF training with that initialization is much slower than MEMM training.Fortunately,we can use the optimal MEMM parameter vector as a starting point for training the corresponding CRF.In Figure5.3, MEMM was trained to convergence in around100iter-ations.Its parameters were then used to initialize the train-ing of CRF,which converged in1,000iterations.In con-trast,training of the same CRF from the uniform distribu-tion had not converged even after2,000iterations.6.Further Aspects of CRFsMany further aspects of CRFs are attractive for applica-tions and deserve further study.In this section we brieﬂy mention just two.Conditional randomﬁelds can be trained using the expo-nential loss objective function used by the AdaBoost algo-rithm(Freund&Schapire,1997).Typically,boosting is applied to classiﬁcation problems with a small,ﬁxed num-ber of classes;applications of boosting to sequence labeling have treated each label as a separate classiﬁcation problem (Abney et al.,1999).However,it is possible to apply the parallel update algorithm of Collins et al.(2000)to op-timize the per-sequence exponential loss.This requires a forward-backward algorithm to compute efﬁciently certain feature expectations,along the lines of Algorithm T,ex-cept that each feature requires a separate set of forward and backward accumulators.Another attractive aspect of CRFs is that one can imple-ment efﬁcient feature selection and feature induction al-gorithms for them.That is,rather than specifying in ad-vance which features of to use,we could start from feature-generating rules and evaluate the beneﬁt of gener-ated features automatically on data.In particular,the fea-ture induction algorithms presented in Della Pietra et al. (1997)can be adapted toﬁt the dynamic programming techniques of conditional randomﬁelds.7.Related Work and ConclusionsAs far as we know,the present work is theﬁrst to combine the beneﬁts of conditional models with the global normal-ization of randomﬁeld models.Other applications of expo-nential models in sequence modeling have either attempted to build generative models(Rosenfeld,1997),which in-volve a hard normalization problem,or adopted local con-ditional models(Berger et al.,1996;Ratnaparkhi,1996; McCallum et al.,2000)that may suffer from label bias. Non-probabilistic local decision models have also been widely used in segmentation and tagging(Brill,1995; Roth,1998;Abney et al.,1999).Because of the computa-tional complexity of global training,these models are only trained to minimize the error of individual label decisions assuming that neighboring labels are correctly -bel bias would be expected to be a problem here too.An alternative approach to discriminative modeling of se-quence labeling is to use a permissive generative model, which can only model local dependencies,to produce a list of candidates,and then use a more global discrimina-tive model to rerank those candidates.This approach is standard in large-vocabulary speech recognition(Schwartz &Austin,1993),and has also been proposed for parsing (Collins,2000).However,these methods fail when the cor-rect output is pruned away in theﬁrst pass.。

lbp3500维修手册

Indicates an item requiring care to avoid electric shocks.
Indicates an item requiring care to avoid combustion (fire).
Indicates an item prohibiting disassembly to avoid electric shocks or problems.
1.3 Product Specifications ................................................................................................................................1- 1 1.3.1 Specifications .......................................................................................................................................................... 1- 1
1.4 Name of Parts.............................................................................................................................................1- 3 1.4.1 External View........................................................................................................................................................... 1- 3 1.4.2 Cross Section .......................................................................................................................................................... 1- 4

惠普彩色激光打印机 Pro M454 和惠普彩色激光多功能一体机 Pro M479 维修手册说明书

Table -1 Revision history Revision number 1
Revision date 6/2019
Revision notes HP LaserJet Pro M454 HP LaserJet Pro MFP M479 Repair manual initial release
Additional service and support for HP internal personnel HP internal personnel, go to one of the following Web-based Interactive Search Engine (WISE) sites: Americas (AMS) – https:///wise/home/ams-enWISE - English – https:///wise/home/ams-esWISE - Spanish – https:///wise/home/ams-ptWISE - Portuguese – https:///wise/home/ams-frWISE - French Asia Pacific / Japan (APJ) ○ https:///wise/home/apj-enWISE - English ○ https:///wise/home/apj-jaWISE - Japanese ○ https:///wise/home/apj-koWISE - Korean ○ https:///wise/home/apj-zh-HansWISE - Chinese (simplified)
Find information about the following topics ● Service manuals ● Service advisories ● Up-to-date control panel message (CPMD) troubleshooting ● Install and configure ● Printer specifications ● Solutions for printer issues and emerging issues ● Remove and replace part instructions and videos ● Warranty and regulatory information

Kidde K85001-0642 火警报警器说明书

Page 1 of 6 D A T A S H E E TK85001-0642Not to be used for installation purposes. Issue 1Kidde Catalog u Speakers, TelephonesTechnology that saves livesStandard Features• High Fidelity 520 Hz speaker models availableLow frequency output meets NFPA standards for newly con-structed commercial sleeping areas; increases sound fidelity and audio intelligibility.• Unique low-profile design– The most compact UL/ULC listed speaker-strobe available – Ultra-slim, protrudes a mere one inch from the wall – Attractive appearance, no visible mounting screws • Field configurable – no need to remove the device – ¼, ½, 1, or 2 watt operation and selectable candelaoutput with convenient switches that remain visible even after the unit is installed • Mass Notification models available with amber lenses • Unparalleled performance– loud 90 dBA output ensures clear, crisp audio – Exclusive FullLight strobe technology produces even light distribution– Precision timing electronics meet tough synchronizing standards for strobes when used with compatible modules – Optional field-configurable temporal strobe output– 25 Vrms and 70 Vrms models available, all supplied with a DC blocking capacitor for audio circuit supervision • Easy to install– Fits all standard 4-inch square electrical boxes with plenty of room behind the signal for extra wire – no extension ring or trim plate needed– #18 - #12 AWG terminals – ideal for long runs or using existing wiringOverviewThe Genesis line of life safety and emergency communications speakers and speaker-strobes combine high performance output with a low profile design to deliver a life safety audio solution that’s as versatile as it is effective. Protruding no more than one inch from the wall, these appliances blend inconspicuously with any decor.Optional amber lens tints, ALERT or FIRE markings, and red or white housing colors ensure there is a device for every application, including mass notification and emergency communications. Speakers feature selectable wattage taps, while speaker-strobes allow for both wattage and light output levels to be configured in the field. Both settings remain clearly visible — even after final installa-tion, which allows devices to be easily fine-tuned to achieve maxi-mum benefit in exchange for the lowest possible system overhead.High fidelity models meet the NPFA 520 Hz requirements for newly construced commercial sleeping areas. They also produce crisp, clear voice audio output that is highly intelligible over large areas.All Genesis speakers include a DC blocking capacitor to allow electrical supervision of the audio distribution circuit. Models for 25 V RMS and 70 V RMS audio circuits are available. With their sealed back construction, these speakers are extra durable and provide outstanding audibility.Wall Speakers, Speaker-StrobesGenesis G4 Series0211/0285See Specifications Section for listings details.High Fidelity ModelsGenesis G4HF Series High Fidelity appliances provide highly intel-ligible voice audio output. They are also effective in areas subjectto high levels of ambient noise. These appliances are approved foruse in sleeping areas under conditions described below.Sleeping Room ApplicationsGenesis G4HF Series High Fidelity appliances are ideal for hotels,dormitories, and other residential occupancies where audibleoutput must meet the 520 Hz signaling characteristics required byNFPA 72.In sleeping areas, always ensure that the wattage tap of thespeaker is set sufficiently high so that the sound pressure reachesat least 75 dBA-fast at the pillow.These appliances are part of an end-to-end audio system ap-proved for use in sleeping areas when used in conjunction withapproved audio hardware and a factory-supplied 520 Hz tone.Check the System Compatibility List for other 520 Hz signalingrequirements.NOTE: Speakers driven by third-party audio systemsare not UL approved for use in sleeping rooms.Genesis mass notificationappliances bring the samehigh-performance life safetyfeatures and unobtrusivedesign to mass notificationapplications. Standard mod-els are available with clear oramber lenses and optionalALERT housing labels, theyare ideal for applicationsthat require differentiation between life safety and mass notificationalerts. Appliances with red, green or blue lenses are available.Field ConfigurationGenesis speakers may be set for ¼, ½, 1, or 2 watt operation. Thewattage setting is visible through a small window on the bottomof the device and is changed by simply sliding the switch until thedesired setting appears in the window. The speaker does not haveto be removed to change the wattage.Genesis speaker-strobes feature selectable candela output. Theoutput setting is visible through a small window on the bottom ofthe device and is changed by simply sliding the switch until thedesired setting appears in the window. The speaker-strobe doesnot have to be removed to change the output.To change strobe to temporalUse the Candela Switch and theWattage switch to set desired operation.Genesis speaker-strobes may also be configured for temporalflash. This battery-saving feature is intended for private modesignaling only. To set the device for temporal flash, snip the circuitboard as shown in the Jumper Locations diagram above.WARNING: These devices will not operate without electrical power.As fires frequently cause power interruptions, we suggest you dis-cuss further safeguards with your local fire protection specialist.Page 2 of 6D A T A S H E E T K85001-0642Not to be used for installation purposes. Issue 1Page 3 of 6 D A T A S H E E TK85001-0642Not to be used for installation purposes. Issue 1Installation and MountingAll models are intended for indoor wall mounted applications only. Speakers and speaker-strobes are flush mounted to a North-American 4” square electrical box, 21/8” (54 mm) deep or a Euro-pean 100 mm square box. Signals may be surface mounted to a Genesis surface-mount box (see ordering information for details).Two tabs at the top of the signal unlock the cover to facilitate mounting. The shallow depth of Genesis devices leaves room behind the signal for extra wiring. Once installed with the cover in place, no mounting screws are visible.Kidde recommends that these speaker-strobes always be installed in accordance with the latest recognized edition of national and lo-cal codes. Refer to installation sheet for mounting height informa-tion.WiringField wiring is connected to Genesis signals with terminals that ac-commodate #18 to #12 AWG (0.75 mm² to 2.5 mm²) wiring.Light outputPer cent of UL rating versus angle11223344556677889910112233445566778899108590Minium UL required candela light output24 Vdc 55 (65)78 (86)153 (159)196 (203)31 Vdc 45 (53)63 (69)120 (124)151 (157)20 Vfwr 56 (106)79 (147)147 (264)197 (342)24 Vfwr 50 (95)68 (130)121 (225)155 (283)27 Vfwr44 (84)60 (115)107 (200)137 (251)Light output switch settings for UL 1971 listed models are selectable by numeric candela value. Light output for Mass Notification (ECS/MNS) appliances is selectable by A, B, C, or D designations.Sound level outputUL 1480: Sound level output at 10 ft (3.05 m) measured in a reverberant room using 400 to 4,000 Hz band limited pink noise. ULC-S541: Sound level output at 10 ft (3.05 m) measured in anechoic chamber using 0 to 4,000 Hz band limited pink noise.G4 Standard Frequency ModelsUL 1480: Sound level output at 10 ft (3.05 m) measured in a reverber-ant room using 400 to 4,000 Hz band limited pink noise.1/2 Watt 83 dBA 1 Watt 86 dBA 2 Watt89 dBAAmber95 cd65 cd26 cd13 cdSpecificationsClear Strobe Output Rating UL 1971: 15 cd (fixed 15/75 cd models)UL 1638, ULCS526: 75 cd (fixed 15/75 cd models)Amber Strobe Output Rating UL 1638: 13 (D), 26 (C), 65 (B), 95 (A)Strobe Operating Voltage16 - 33 Vdc Regulated, 16-33 V Full wave rectified (UL Voltage Designations “Regulated 24” and “24 fwr”) Strobe Flash Rate One flash per second.Strobe Flash Synchronization All strobes: one flash per second (fps) within 200 milliseconds over 30 minutes on common circuit. All strobes: Synchronization source required to comply with UL 1971 synchronization standard. Temporal setting (private mode only): synchronized to temporal output on the same circuit.Synchronization Sources FACPs: VM and VS Series life safety systems, FX Series fire alarm control panels. Moduels: GSA-CC1S, GSA-MCC1S, SIGA-CC2A, GSA-MCC2A, EG1M-RM. Power supplies: MIRBPS6A, MIRBPS10A, APS6A, APS10A.Strobe Lens Material PolycarbonatePage 4 of 6D A T A S H E E T K85001-0642Not to be used for installation purposes. Issue 1Page 5 of 6 D A T A S H E E TK85001-0642Not to be used for installation purposes. Issue 1Ordering InformationLife safety Appliances G4-S2WhiteNoneNoneSpeaker only models25 Volt (Selectable ¼, ½, 1, or 2watt)1.5 lbs. (0.68 kg)G4HFWN-S2üG4R-S2 RedG4HFRN-S2üG4F-S2WhiteFIREG4HFWF-S2üG4RF-S2 Red G4HFRF-S2üEG4-S2VMWhiteNoneClearSelectable 15, 30, 75, or 110 cdG4HFWN-S2VMC üEG4R-S2VMRedG4HFRN-S2VMC üEG4F-S2VMWhiteFIREG4HFWF-S2VMC üEG4RF-S2VM Red G4HFRF-S2VMC üG4-S7WhiteNoneNoneSpeaker only models70 V (Selectable ¼, ½, 1, or 2watt)G4HFWN-S7üG4R-S7RedG4HFRN-S7üG4F-S7WhiteFIREG4HFWF-S7üG4RF-S7Red G4HFRF-S7üEG4-S7VMWhiteNoneClearSelectable 15, 30, 75, or 110 cdG4HFWN-S7VMC üEG4R-S7VMRed G4HFRN-S7VMC üEG4F-S7VMWhite FIRE G4HFWF-S7VMC üG4HFRF-S7VMC üRed EG4F-S7V1575White FIREClear15/75 cd¹EG4RF-S7V1575RedMass Notification Appliances G4WA-S2VMA*WhiteALERTAmberSelectable 13, 26, 65, or 95 cd 25 Volt (Selectable ¼, ½, 1, or 2watt)1.5 lbs. (0.68 kg)G4HFWA-S2VMA*üG4WA-S2VMC ClearSelectable15, 30, 75, or 110 cd G4HFWA-S2VMC üG4WN-S2VMA*NoneAmber Selectable 13, 26, 65, or 95 cd G4HFWN-S2VMA*üG4WN-S2VMC Clear15, 30, 75, or 110 cdG4WA-S2ALERT NoneSpeaker only models G4HFWA-S2üG4WN-S2NoneG4WA-S7VMA*WhiteALERTAmberSelectable 13, 26, 65, or 95 cd 70 V (Selectable ¼, ½, 1, or 2watt)G4HFWA-S7VMA*üG4WA-S7VMC ClearSelectable15, 30, 75, or 110 cd G4HFWA-S7VMC üG4WN-S7VMA*NoneAmber Selectable 13, 26, 65, or 95 cd G4HFWN-S7VMA*üG4WN-S7VMC Clear15, 30, 75, or 110 cdG4WA-S7ALERT None Speaker only modelsG4HFWA-S7üG4WN-S7None* Not approved for fire alarm applicationsTechnology that saves livesContact us...Email:*****************.comWeb: /EngineeredSystems Kidde is a UTC brand.1016 Corporate Park DriveMebane, NC 27302© 2016 United Technologies Corporation. All rights reserved.AccessoriesEG1M-RM Synchronization Output Module (1-gang)0.2 (0.1) GSA-CC1S Intelligent Synchronization Output Module (2-gang)0.5 (0.23) GSA-MCC1S Synchronization Output Module (Plug-in UIO)0.18 (0.08) EG4B Surface mount box, white0.7 (0.32) EG4RB Surface mount box, red0.7 (0.32)Page 6 of 6D A T A S H E E T K85001-0642Not to be used for installation purposes. Issue 110-11-16。

New_Energy_Vehicle_Market_Accelerates_the_Combing-

60In April 2022, BYD off icially announced the decision to stop the production of fuel vehicles, meaning it has completely moved away from the “fuel era”, and will focus on pure electric and plug-in hybrid electric vehicles in the future. In other words, BYD will concentrate its resources on strengthening the competitiveness in new energy vehicle (NEV) market.The pandemic causes changes in industry landscapeEntering 2022, the COVID-19 pandemic broke out again in China, c a u s i n g s e v e r a l p r o v i n c e s a n d municipalities, including Shanghai and Jilin Province, to be under full or partial regional lockdown. The outbreak has had a serious impact on both the supply and demand sides of China’s automobile industry, resulting in a significant decline in domestic vehicle sales in April. According to the data from China Passenger Car Association (CPCA), domestic wholesale volume of narrowly-defined passenger vehicles fell 43.0% year-on-year in April and 47.8% from March, while retail sales fell 35.5% and 34.0% respectively withNew Energy Vehicle Market Accelerates the Combing-out of Unfit Playersthe same comparisons.In May, the epidemic in China becomes gradually under control, as FAW Group, SAIC Motor, Tesla and other vehicle producers and supporting parts enterprises resume production, and the tight supply begins to ease. However, the epidemic would have an long-lasting impact on the demand side, especially for NEV sector, which continued to grow at a high rate before the outbreak. After current wave of the epidemic quiets down, there is expected to be a significant change in the landscape of new energy vehicle and parts industry, comparing to that before the epidemic.According to sources in the industr y, before the outbreak of COVID-19 in 2019, many domestic b r a n d s a n d n e w l y-e s t a b l i s h e d automobile companies in China were able to occupy a certain share in new energy passenger vehicle market, and some of them with poor product competitiveness gradually lost their market shares between 2018 and 2021, and even face a survival crisis.In 2018, all the top ten players (in terms of market share) in new energy passenger vehicle market in China were domestic enterprises, with BYD, BAIC BJEV, SAIC Motors, Chery and JAC Motor of the top five. In 2019, most of the top ten players were still domestic automobile companies, with BYD and BAIC BJEV holding basically the same market share in 2019 as in 2018, and still taking the lead, while small and medium-sized brands, such as Huatai Auto, JMEV and Zotye Automobile whose products and brands are less competitive, disappeared from the top ten in 2019 and were replaced by SAIC-GM-Wuling, GAC Aian, Great Wall Motor, SAIC-VW and other leading domestic automobile companies.In 2021, the new energy passenger vehicle market in China was basically stable, since the top ten players compared to those in 2020 did not change much, but new players of NIO, Xiaopeng, and Lixiang were enlisted among the top ten.Analysts believe that since 2021, NEV market has kept developing fast, and all major NEV enterprises have benefited from rapidly-growing market. It is estimated that 3 to 5 years later when these favorable policies terminate, and the market enters the stage of stock competition, the growth rate of NEV sales may slow down, and NEV enterprises with weak competitiveness will gradually be eliminated.In the post-epidemic era, while high-end brands with strong product and brand competitiveness, such as Tesla, will maintain their advantages in middle and high-end market (for vehicles priced above RMB 200,000), the declining purchasing power of low-income consumers will lead to more intense competition in middle and low-end market, and automobile companies with leading product technology and cost performance, such as BYD, will quickly increase their presence in middle and low-end market, at which they are targeting. It is expected that the above two types of automobile companies will continue to increase their market shares, and some relatively less competitive players will have to face greater survival crisis, as low-income consumers have weaker demand, and Huawei, Xiaomi and other technology companies are entering NEV market from 2023 onwards, putting higher pressure on existing players.More players are entering NEV marketMore strong players flooded inBy Jenny Hu61new energy passenger vehicle market in China in 2020, which witnessed a major change in new energy passenger vehicle industry, as Tesla China, NIO, FAW-VW, and Lixiang be enlisted among the top ten players in this market in China, and SGMW capture the mid- and low-end market with new models such as Hongguang MINI EV, bringing its market share from 4.5% in 2019 to 14.8% in 2020. At the same time, BAIC BJEV, who held the largest market share in 2019, almost disappeared in 2020, with sales slipping from nearly 150,000 vehicles in 2019 to 25,900 in 2020.From 2022 onwards, Huawei, Xiaomi, Baidu and other technology companies will further intensify the competition in new energy passenger vehicle market, putting greater pressure on automobile companies with weaker product and brand competitiveness, and accelerating the combing-out process in NEV industry.Huawei announced the establishment of a business unit for intelligent automobile solutions in 2019, and has now launched self-developed products and solutions in fields such as intelligent driving, intelligent cockpit, core systems (battery, electric motor and electronic control system) for NEVs, thermal management, etc., which have been applied in several flagship models from intelligent automobile brands such as ArcFox, Avatr and AITO.On March 30, 2021, Xiaomi made an announcement at Hong Kong Stock Exchange, and officially announced its entry into intelligent electric vehicle industry. It’s learned from Xiaomi’s financial report 2021, now its automobile business department has an R&D team of exceeding 1,000 employees, and the f irst model is expected to be mass-produced in the first half of 2024. In March 2021, Baidu and Geely announced the establishment of Jidu Automobile, to take each advantages and research and develop intelligent connected vehicles: Baidu will provide Jidu with intelligent software capabilities such as autonomous driving and intelligent connection, while Geely will provide SE A (S u s t a i n a b l e E x p e r i e n c e Architecture) and complete vehiclemanufacturing capabilities.In recent years BYD made a series of strategic adjustments, completed the shift from vertical integration to an open platform, and developed a fully open supply chain, which will benefit NEV parts industry chain and help BYD enhance its brand and influence through the industry. In 2021 BYD made great achievements in both pure electric and hybrid fields, becoming a leader in new energy passenger vehicle market and taking the first place in terms of sales of new energy passenger vehicles in China. Besides, BYD became the first domestic brand that realized sales of 100,000 new energy passenger vehicles per month, succeeding in being enlisted among head companies in passenger vehicle market in China.Fiercer competition in high-end NEV marketIn last three or four years, a small number of automobile companies, such as NIO, Lixiang and Xiaopeng, have actively explored high-end NEV market and made remarkable progresses, with sales approaching 100,000 vehicles in 2021 respectively, inspiring many other players to accelerate the layout for high-end NEV market. Entering 2022, following Lixiang, NIO and Xiaopeng, a large number of new high-end NEV brands, including Shalong Auto, Avatr, IM Motors, Dongfeng M series, AITO, and Niutron, are actually advancing their production plans.According to a survey by Yiche Research, in 2021 the consumers in high-end NEV market in China were mainly from sectors of manufacturing, Internet, energy, construction and so on. The rise of some high-end NEV brands is closely related to consumers from Internet-related industries, as the survey indicated that nearly two out of ten (18.64%) such buyers chose NEVs in 2021.Also thanks to the contribution from consumers in Internet-related industries, high-end NEV brands such as Lixiang and NIO develop quickly. In 2021, Lixiang and NIO led in China in terms of the penetration among consumers in Internet-relatedindustries, reaching 31.18% and 18.18% respectively, following by AIAN, Tesla and Xiaopeng, fully demonst rat ing t hat (h igh-end) NEV brands that focus on product innovation and users’ preferences are more attractive to such consumers. The fact that Link and Tank are among top 10 in terms of brand penetration indicates consumers in Internet-related industries prefer individuality and segmented brands, and that BMW, Mercedes-Benz, Audi, Porsche, Lexus and other high-end brands are among top 20 shows that these consumers have stronger consuming power.In addition, the survey showed that women were of a force quite noticeable in automobile market for consumers from Internet-related industries. In 2021, although most such automobile consumers were still male, women also made up nearly 40%. At the same time, up to 5.59% of female automobile buyers were from Internet-related industries, which was significantly higher that of male. High-end NEV brands need an active layout plan for the female automobile market, to attract consumers from Internet-related industries and achieve a rapid growth, by specifically optimize model design, sale network distribution, after-sales service, etc.However, analysts believe there are signs that more and more high-end NEV brands are getting homogenized, including products and ser vices, and they are lack of innovative ideas attractive to consumers from Internet-related industries. During the 14th Five-Year Plan period, if high-end NEV brands want to attract these consumers and secure a rapid growth in sales, they must develop innovative corporate culture, and strive to building a differentiated competition system from Lixiang, NIO, and so on, to better attract quality consumers. As for specific strategies, they may need to focus on the four first-tier cities in China of Beijing, Shanghai, Guangzhou and Shenzhen, and actively put medium and medium-to-large-sized, sports cars and other personalized models, to meet the demand for add-on or replacement purchase.。

Chem. Commun. 2012, 48, 7477-7479

Cite this:mun .,2012,48,7477–7479Guest-induced self-assembly of a macrocyclic boronic ester containing diarylethene units:enhancement of photoresponsivity wNobuharu Iwasawa,*a Hiroki Takahagi,a Kosuke Ono,a Kotaro Fujii b and Hidehiro Uekusa bReceived 19th March 2012,Accepted 11th June 2012DOI:10.1039/c2cc31989fGuest-induced self-assembly of a macrocyclic boronic ester containing photochromic diarylethene units is realized and this macrocycle showed high quantum yield of photoisomerization due to favourable conformational constraint.Macrocyclic host molecules containing photochromic units are an interesting class of compounds,which can act as basic frameworks for unique molecular switches,etc.1Diarylethenes,in particular,dithienylethenes are one of the most successful photochromic molecules that undergo a photochemical open/closed-ring isomerization,resulting in large changes in their electronic and steric properties.2Photocyclization involving 6p electrons can proceed only from the antiparallel conformer of open-isomers,while the parallel conformer is photochemically inactive according to Woodward–Hoﬀman rules.In contrast to the numerous studies on the monomeric or linear oligo-or polymeric diarylethenes,2macrocyclic architecture of diarylethenes that would open up a unique photoresponsive host–guest chemistry has been explored much less frequently due to the diﬃculty in designing and combining appropriate component molecules.3Therefore,it is an attractive challenge to realize self-assembly of macrocyclic host molecules containing diarylethene units not only from the stand-point of the diarylethene chemistry but also for the utilization of such molecules in molecular devices;however,to our knowledge,there has been no example of guest-induced self-assembly of macrocyclic host molecules bearing diarylethene units.3We have already reported the guest-induced dynamic self-assembly of two kinds of macrocyclic boronic esters,in which selective precipitation of macrocyclic boronic esters occurs by a simple mixing of 1,4-benzenediboronic acid and indacene bis(1,2-diol)in the presence of appropriate guest molecules.4In this paper,we report realization of a highly selective self-assembly of a macrocyclic boronic ester containing two diarylethene units,which has realized high quantum yield of photoisomerization dueto the favorable constraint of the conformation of the mole-cule by macrocycle formation.We ﬁrst examined the self-assembly of macrocyclic boronic esters by mixing equimolar amounts of the open-form diarylethene unit 15and the racemic bis(1,2-diol)2containing an indacene framework 4in MeOH in the presence of various guest molecules.Only insoluble polymeric boronic esters were obtained in almost all cases;5however,when equimolar amounts of 1and 2were mixed in MeOH–benzene (2:1),white solids 3,which were soluble in several organic solvents and found to be composed of one molecule each of 1,2,and benzene,precipitated out in 81%yield.FAB-MASS analysis suggested the formation of a macrocyclic boronic ester 3containing two molecules each of 1and 2,and thus two molecules of benzene were included in the precipitated 3(Scheme 1).The structure of 3was successfully determined by X-ray crystallographic analysis (Fig.1).5A single crystal,suitable for X-ray diﬀraction analysis,was obtained from a toluene solution of 3.As shown in Fig.1a,this macrocyclic boronic ester located on a crystallographic 2-fold rotation axis.Scheme 1Self-assembly of macrocyclic boronic ester 3containing diarylethene units from 1and rac -2.Fig.1Crystal structures of 3Á6toluene.(a)Stick representation of host 3.The toluene molecules are omitted for clarity.The dotted blue line represents the 2-fold rotation axis.(b)Space-ﬁlling representation of 3Á6toluene.Two toluene molecules which exist in the host molecule are shown in orange.The other toluene molecules are omitted for clarity.Department of Chemistry,Tokyo Institute of Technology and CREST,Japan Science and Technology Agency (JST),2-12-1,O-okayama,Meguro-ku,Tokyo 152-8551,Japan.E-mail:niwasawa@chem.titech.ac.jp;Fax:+81357342931bDepartment of Chemistry and Materials Science,Tokyo Institute of Technology and CREST,Japan Science and Technology Agency (JST),2-12-1,O-okayama,Meguro-ku,Tokyo 152-8551,Japan w Electronic supplementary information (ESI)available:Experimental details and additional DC 825696and 847902.For ESI and crystallographic data in CIF or other electronic format see DOI:10.1039/c2cc31989fChemCommDynamic Article Links/chemcommCOMMUNICATIOND o w n l o a d e d b y S o o c h o w U n i v e r s i t y C h i n a o n 06 S e p t e m b e r 2012P u b l i s h e d o n 13 J u n e 2012 o n h t t p ://p u b s .r s c .o r g | d o i :10.1039/C 2C C 31989FView Online / Journal Homepage / Table of Contents for this issueClearly,the macrocyclic boronic ester is a homo compound composed of the same two enantiomers of tetrol 2and has a highly symmetrical structure,and both the diarylethene units take the expected antiparallel conformation which would be suitable for photoisomerization.6Furthermore,six toluene mole-cules were incorporated in the crystal,two of which existed in the host molecule and the remaining in the intermolecular interstices (Fig.1b).Although the quality of the data was low,X-ray analysis of crystals of 3,obtained by vapor diﬀusion of hexane into a benzene solution of 3,also showed two molecules of benzene inside the cage,5suggesting that benzene is likely to act as a template to induce the self-assembly of 3from 1and 2.7The 1H NMR spectrum of the single crystals of 3in CD 2Cl 2(Fig.2b)indicated the presence of two conformational isomers concerning the diarylethene moieties in a 95:5ratio.8Furthermore,the ratio of the two conformers changed to 87:13in d 6-DMSO,suggesting that these two are the rotational isomers in rapid equilibrium.Careful examination of the molecular models suggests that there would be 6possible conformational isomers distinguishable by 1H NMR for 3,that is,three antiparallel–antiparallel isomers shown in Fig.2a and additional three parallel–parallel isomers.5,9Calculation at DFT level revealed that 3major shown in Fig.2a,which is actually the same as that disclosed by X-ray analysis,is the most stable conformer in solution and the other two antiparallel–antiparallel conformers 3minor and 3other are about 2–3kcal mol À1less stable.5The 1H NMR spectrum shown in Fig.2b suggests that both isomers observed are highly symmetrical,and it is supposed that the minor isomer observed would be 3minor shown in Fig.2a although the calculated energy value of 3minor is slightly higher than that of 3other .10It should be noted that in contrast to the monomeric diarylethene pinacolato boronic ester of 1,which consists of both parallel and antiparallel isomers in about 43:57ratio in CD 2Cl 2,5both the diaryl-ethene units in the observed two macrocyclic boronic esters 3were thought to take antiparallel conformation based on theirhigh symmetry and their analogous chemical shift to that of the methyl group on the thienyl ring.5,11The photoisomerization behavior of the macrocyclic boronic ester 3was monitored by UV-vis spectroscopy.Upon irradiating a CH 2Cl 2solution of 3with UV light of 313nm,new bands appeared at 365and 550nm,corresponding to closed-ring isomers (Fig.3).Several clear isosbestic points conﬁrmed that the two diarylethene units in the macrocyclic boronic ester behaved independently without obvious deviation from the other diarylethene unit.The 1H NMR spectrum of the irradiated mixture suggested the formation of a major product along with several minor products as shown in Fig.4b (top).Analysis of the structure of these products is not necessarily easy,as a maximum of three diastereomeric isomers 4a–c ,which correspond to the three antiparallel–antiparallel isomers of 3,12would be possible for the product in which both the diarylethene units are ring-closed.However,precipitation ofFig.2(a)Structures of three conformational isomers of 3.The values in parentheses are the calculated relative energies.(b)1H NMR spectrum (400MHz,CD 2Cl 2)of macrocyclic boronic esters 3(K :minor conformational isomer of 3).Fig.3UV/vis absorption spectral changes of 3(2.0Â10À5M)in CH 2Cl 2upon UV light irradiation (313nm)at 0,20,60,100,180,270and 1200s.Isosbestic points are observed at 256,282,305,and 321nm.The inset shows time-dependent absorption change at 550nm.Fig.4(a)Possible diastereomers of closed-ring form of macrocyclic boronic esters 4.(b)1H NMR spectra (400MHz,CD 2Cl 2)of photo-stationary state (top)and isolated closed-form 4(bottom).D o w n l o a d e d b y S o o c h o w U n i v e r s i t y C h i n a o n 06 S e p t e m b e r 2012P u b l i s h e d o n 13 J u n e 2012 o n h t t p ://p u b s .r s c .o r g | d o i :10.1039/C 2C C 31989Fsolids occurred during irradiation at a higher concentration,and careful isolation and 1H NMR measurement of this precipitated product revealed that this is the major product observed in the crude product (Fig.4b,bottom).Although we have not yet succeeded in obtaining a single crystal suitable for X-ray analysis,the major product is assumed to be 4c ,shown in Fig.4a,because 1H NMR showed two sets of signals of nearly equal intensity as shown in Fig.4b suggesting lower symmetry of the product.12Actually 4c is not obtained from either of the antiparallel–antiparallel isomers 3major or 3minor shown in Fig.2a without conformational change.As the photocyclization proceeds much faster than the conformational change,the ﬁrst cycliza-tion should occur from 3major ,and it is supposed that after the ﬁrst cyclization occurred from 3major ,conformational change had occurred at the remaining open-form diarylethene moiety to give 4c after the second cyclization.Calculation of the relative energy of the possible mono-cyclized intermediates supports this possibility.5Finally,we measured the photoresponsivity of this cycloiso-merization.And it was a great pleasure to ﬁnd that a quantum yield of 0.67for ring-closing photoisomerization,which was more than double the value of the corresponding monomeric pinacolato boronic ester of 1(f =0.30),was realized by the initial rate method.This value of 0.67is among the highest quantum yields for the photoisomerization of non-polymeric diarylethene derivatives.2,13The favorable constraint of the con-formation of the diarylethene moiety to the antiparallel conformer by the macrocycle formation has enhanced the photoresponsive eﬃciency.The cycloreversion quantum yield value of 0.64was also higher and more than twice that of the closed form of 1(f =0.27).This possibly originates from the strain of the closed-ring form of macrocyclic boronic esters 4.In conclusion,we have succeeded in realizing selective guest-induced self-assembly of the macrocyclic boronic ester containing photochromic diarylethene units.Noteworthy is a very easy construction of the deﬁnite macrocyclic boronic ester simply by mixing two components in MeOH with benzene as the guest molecule,and this macrocycle showed a high quantum yield of photoisomerization due to the favorable conformational constraint.We are now trying to construct photoresponsive macrocyclic boronic esters,which could be utilized in a molecular sensor or a molecular switch.This work was supported by a CREST (Core Research for Evolution Science and Technology)project from the Japan Science and Technology Agency (JST).Notes and references1For representative examples,see:(a )S.Shinkai,T.Ogawa,T.Nakaji,Y.Kusano and O.Manabe,Tetrahedron Lett.,1979,20,4569–4572;(b )A.Ueno,H.Yoshimura,R.Saka and T.Osa,J.Am.Chem.Soc.,1979,101,2779–2780;(c )I.Yamashita,M.Fujii,T.Kaneda,S.Misumi and T.Otsubo,Tetrahedron Lett.,1980,21,541–544;(d )M.Asakawa,P.R.Ashton,V.Balzani,C.L.Brown,A.Credi,O.A.Matthews,S.P.Newton,F.M.Raymo,A.N.Shipway,N.Spencer,A.Quick,J.F.Stoddart,A.J.P.White and D.J.Williams,Chem.–Eur.J.,1999,5,860–875;(e )T.Muraoka,K.Kinbara and T.Aida,Nature ,2006,440,512–515;(f )H.-S.Tang,N.Zhu and V.W.-W.Yam,Organometallics ,2007,26,22–25;(g )M.Liu,X.Yan,M.Hu,X.Chen,M.Zhang,B.Zheng,X.Hu,S.Shao and F.Huang,Org.Lett.,2010,12,2558–2561;(h )M.Yamamura,Y.Okazaki and T.Nabeshima,mun.,2012,48,5724–5726.2(a )M.Irie,Chem.Rev.,2000,100,1685–1716;(b )H.Tian and S.Yang,Chem.Soc.Rev.,2004,33,85–97;(c )M.Irie,Proc.Jpn.Acad.,Ser.B ,2010,86,472–483;(d )H.Dong,H.Zhu,Q.Meng,X.Gong and W.Hu,Chem.Soc.Rev.,2012,41,1754–1808.3There are several examples for the preparation of macrocyclic compounds containing diarylethenes utilizing coordination or covalent bond formation.See for examples,(a ) E.Murguly,T.B.Norsten and N.R.Branda,Angew.Chem.,Int.Ed.,2001,40,1752–1755;(b )B.Qin,R.Yao,X.Zhao and H.Tian,Org.Biomol.Chem.,2003,1,2187–2191;(c )I.Jung,H.Choi,E.Kim,C.-H.Lee,S.O.Kang and J.Ko,Tetrahedron ,2005,61,12256–12263;(d )J.Yin,Y.Lin,X.Cao,G.-A.Yu and S.H.Liu,Tetrahedron Lett.,2008,49,1582–1585;(e )J.Ka rnbratt,M.Hammarson,S.Li,H.L.Anderson, B.Albinsson andJ.Andreasson,Angew.Chem.,Int.Ed.,2010,49,1854–1857.4(a )N.Iwasawa and H.Takahagi,J.Am.Chem.Soc.,2007,129,7754–7755;(b )H.Takahagi,S.Fujibe and N.Iwasawa,Chem.–Eur.J.,2009,15,13327–13330;(c )H.Takahagi and N.Iwasawa,Chem.–Eur.J.,2010,16,13680–13688.See also:(d )N.Fujita,S.Shinkai and T.D.James,Chem.–Asian J.,2008,3,1076–1091.5For details,see the ESI w .6This does not necessarily mean that the major conformational isomer in solution has the same conformation as that obtained by this X-ray analysis;however,we believe that they take the sameconformation in solid and in solution as described in a later section.7Concerning the selective preparation of 3Á2benzene,there is another possibility that solubility of the product plays an impor-tant role in this system.At present we believe that benzene works as a guest molecule to induce the precipitation of 3.For the importance of the precipitation process in the macrocyclic boronic ester formation,see ref.4.8Single crystals and the precipitates obtained by mixing 1and 2in MeOH–benzene showed the same 1H NMR spectra except for the guest molecules.9Antiparallel–parallel isomers were excluded as they did not seem to be feasible according to molecular models.10The remaining antiparallel–antiparallel conformer 3other has only one C 2axis,which should show 6peaks in the methyl signal region (1–2.5ppm).The possibility of this isomer being the observed minor isomer cannot be completely ruled out because overlapping of the small signals with the major signals might be possible.11(a )K.Uchida,E.Tsuchida,Y.Aoi,S.Nakamura and M.Irie,Chem.Lett.,1999,63–64;(b )S.Nakamura,S.Yokojima,K.Uchida,T.Tsujioka,A.Goldberg,A.Murakami,K.Shinoda,M.Mikami,T.Kobayashi,S.Kobatake,K.Matsuda and M.Irie,J.Photochem.Photobiol.,A ,2008,200,10–18.124a ,4b ,and 4c correspond to 3major ,3minor and 3other shown in Fig.2a,respectively.We cannot completely rule out the possibility that the product is an approximate 1:1mixture of 4a and 4b derived from the observed,two highly symmetrical isomers of 3.13There are several examples that show more than 0.60value of quantum yield of cycloisomerization in solution.See for examples:(a )F.Stellacci,M.C.Gallazzi and G.Zotti,Adv.Mater.,1999,11,292–295;(b )M.Takeshita,M.Nagai and T.Yamamoto,mun.,2003,1496–1497;(c )K.Matsuda,Y.Shinkai,T.Yamaguchi,K.Nomiyama,M.Isayama and M.Irie,Chem.Lett.,2003,32,1178–1179;(d )S.Kawai,T.Nakashima,Y.Kutsunugi,H.Nakagawa,H.Nakano and T.Kawai,J.Mater.Chem.,2009,19,3606–3611.See also,ref.3c and (e )K.Morinaka,T.Ubukata and Y.Yokoyama,Org.Lett.,2009,11,3890–3893;(f )S.Aloıse,M.Sliwa,Z.Pawlowska,J.Rehault,J.Dubois,O.Poizat,G.Buntinx, A.Perrier,F.Maurel,S.Yamaguchi and M.Takeshita,J.Am.Chem.Soc.,2010,132,7379–7390;(g )S.Fukumoto,T.Nakashima and T.Kawai,Angew.Chem.,Int.Ed.,2011,50,1565–1568.D o w n l o a d e d b y S o o c h o w U n i v e r s i t y C h i n a o n 06 S e p t e m b e r 2012P u b l i s h e d o n 13 J u n e 2012 o n h t t p ://p u b s .r s c .o r g | d o i :10.1039/C 2C C 31989F。

ppt课件-l e c t u r e e2( l e c t u r e2)

Chemistry Sources (remember what HCOand Hdo)
Copyright ©2010 R. R. Dickerson
3
HYDROCARBRONS REACTIVITY
FOR URBAN SMOG (OZONE)
FORMATION
HYDROCARBON
k(O) ( A l l u n i t s : cm³s ¹)
Copyright ©2010 R. R. Dickerson
16
Correlation plot of all LIF OHd at a versus the photolysis frequency of ozone, j(O1D). (Adapted from Holland
et al., 1998).
Anderson put a scanning spectrometer on t h e nose of a rocket and measured the emission a t 308 nm due t o s o l a r excitation.
In situ resonance fluorescence with a microwave discharge lamp worked i n t h e s t r a t
LECTURE 20 Atmospheric Odd Hydrogen, HOx
AOSC 637 Spring 2010 Atmospheric Chemistry Russell R. Dickerson
Copyright ©2010 R. R. Dickerson
1
Odd Hydrogen: O u t l i n e

1、下载文档前请自行甄别文档内容的完整性，平台不提供额外的编辑、内容补充、找答案等附加服务。
2、"仅部分预览"的文档,不可在线预览部分如存在完整性等问题,可反馈申请退款(可完整预览的文档不适用该条件!)。
3、如文档侵犯您的权益，请联系客服反馈,我们会尽快为您处理(人工客服工作时间：9:00-18:30)。

a rXiv:081.1122v1[astro-ph]7Jan28Models for Low-Mass X-Ray Binaries in the Elliptical Galaxies NGC3379and NGC4278:Comparison with Observations T.Fragos 1,V.Kalogera 1,K.Belczynski 2,G.Fabbiano 3,D.-W.Kim 3,N.J.Brassington 3,L.Angelini 4,R.L.Davies 5,J.S.Gallagher 6,A.R.King 7,S.Pellegrini 8,G.Trinchieri 9,S.E.Zepf 10,A.Zezas 3tassosfragos@,vicky@,kbelczyn@,gfabbiano@,kim@,nbrassington@,angelini@,rld@,jsg@,ark@,silvia.pellegrini@unibo.it,ginevra.trinchieri@brera.inaf.it,zepf@,azezas@ ABSTRACT We present theoretical models for the formation and evolution of populations of low-mass X-ray binaries (LMXB)in the two elliptical galaxies NGC 3379and NGC 4278.The models are calculated with the recently updated StarTrack code (Belczynski et al.2007),assuming only a primordial galactic ﬁeld LMXB popu-lation.StarTrack is an advanced population synthesis code that has been testedand calibrated using detailed binary star calculations and incorporates all the im-portant physical processes of binary evolution.The simulations are targeted tomodeling and understanding the origin of the X-ray luminosity functions(XLF)of point sources in these galaxies.For theﬁrst time we explore the populationXLF down to luminosities of3×1036erg s−1,as probed by the most recent ob-servational results(Kim et al.2006a).We consider models for the formationand evolution of LMXBs in galacticﬁelds with diﬀerent CE eﬃciencies,stellarwind prescriptions,magnetic braking laws and initial mass functions.We iden-tify models that produce an XLF in excellent agreement with the observationsboth in shape and absolute normalization.We alsoﬁnd that the treatment of theoutburst luminosity of transient systems remains a crucial factor for the deter-mination of the XLF since the modeled populations are dominated by transientX-ray systems.Subject headings:Stars:Binaries:Close,Stars:Evolution,X-rays:Binaries,Galaxies:Ellipticals1.INTRODUCTIONA Low Mass X-ray binary(LMXB)is a Roche lobe overﬂowing,mass-transfering binary system with a compact object accretor,either a black hole(BH)or a neutron star(NS),and a low mass( 1M⊙)donor.Since the late80’s it has been suggested that LMXBs should exist in early type galaxies,E and S0,and that they might even dominate the X-ray emission (Trinchieri and Fabbiano1985;Fabbiano1989;Kim et al.1992).The stellar populations in these galaxies are typically old and homogeneous.Massive stars have already evolved to compact objects and LMXBs are probably the only sources with X-ray luminosities above 1036erg s−1.Uncontroversial detection of LMXBs in early type galaxies became possible only this last decade with Chandra’s increased angular resolution(Fabbiano2006;Sarazin et al.2000). The spectra of individual X-ray sources are consistent with those expected from LMXB models and the LMXBs observed in the Milky-way and M31(Humphrey and Buote2006; Irwin et al.2003).For many galaxies observed with Chandra,the XLFs have been derived and they can usually beﬁtted with a single or a broken power law.The detections limit for these surveys is usually a few times1037erg s−1.Kim and Fabbiano(2004)derived XLFs for 14early type galaxies and they included completeness corrections.Each XLF is wellﬁtted with a single power law with cumulative slope between-0.8and-1.2.The composite XLF of these galaxies though is not consistent with a single power law.There is a prominent breakat(5±1.6)×1038erg s−1,close to the Eddington luminosity(L Edd)of a helium accreting NS-LMXB.This break might be hidden in the individual XLFs due to poor statistics(see also Sarazin et al.2000;Kundu et al.2002;Jord´a n et al.2004;Gilfanov2004).Other recent studies by Jeltema et al.(2003),Sivakoﬀet al.(2003)and Jord´a n et al.(2004)suggested a break of the XLF at a higher luminosity(∼1039erg s−1).The exact position and the nature of these breaks are still somewhat controversial,as the correct interpretation of the observed XLFs relies signiﬁcantly on the proper completeness correction when looking at luminosities close to the detection limit and small number statistics at the high-end of the XLF.Recent Chandra observations(Kim et al.2006a)have yielded theﬁrst low-luminosity XLFs of LMXBs for two typical old elliptical galaxies,NGC3379and NGC4278.The detec-tion limit in these observations is∼3×1036erg s−1which is about an order of magnitude lower than in most previous surveys of early type galaxies.The observed XLFs of the two ellipticals extend only up to6×1038erg s−1and are well represented by a single power law with a slope(in a diﬀerential form)of1.9±0.1.When Chandra observations are compared with optical images from Hubble or other ground based telescopes,it is generally found that a signiﬁcant fraction of the LMXBs are inside globular clusters(GC).On average4-5%of the GCs in a given galaxy are associated with a LMXB(with L x>∼1037erg s−1),while the fraction of LMXBs located in GC,varies from10%to70%depending on the type of the galaxy and its GC speciﬁc frequency.At present the origin and the properties of these systems,both in GCs and theﬁeld,are not yet well understood.It has been noted that XLFs at high luminosities for each sub-group (GCs andﬁeld)do not reveal any diﬀerences within the statistics of the samples consid-ered(Fabbiano2006;Kim et al.2006b;Kundu et al.2007;Jord´a n et al.2004;Sarazin et al. 2003).However,in more recent studies,(Fabbiano et al.2007)and Voss and Gilfanov(2007) independently found that the two XLFs(GCs andﬁeld sources)show signiﬁcant diﬀerences at low luminisities(below1037erg sec−1),pointing to a diﬀerent LMXB formation mechanism in GCs.The natural question that arises is whether(i)all LMXBs were formed in GCs through dynamical interactions and some eventually escaped or some GCs dissolved in theﬁeld, or(ii)ﬁeld LMXBs were born in situ through binary evolution of primordial binaries.The formation rates associated with these two possibilities are not understood well enough to give accurate predictions and use the relative numbers in the samples.Juett(2005)has shown that the observed relationship between the fraction of LMXBs found in GCs and the GC-speciﬁc frequency in early-type galaxies is consistent with the galacticﬁeld LMXB population being formed in situ.Similarly,Irwin(2005)compared the summed X-ray luminosity of the LMXBs to the number of GCs in a galaxy;in the case of all LMXBs having formed exclusivelyin GCs,the two should be directly proportional regardless of where the LMXBs currently reside.Instead he found that the proportionality includes an additive oﬀset implying the existence of a LMXB population unrelated to GCs.In the past,semi-analytical theoretical models have been introduced for the study of the LMXB population in early galaxies.White and Ghosh(1998)studied the connection between the star formation rates of normal galaxies,i.e.galaxies without an active nucleus, and the formation rate of LMXBs and millisecond pulsars,assuming that all LMXBs are formed from primordial binaries.Considering a time-dependent star formation rate,they showed that the general relativity timescales relevant to the evolution of primordial binaries to LMXBs and to millisecond pulsars,lead to a signiﬁcant time delay of the peak in the formation rate of these populations after the peak in the star formation rate.In a followup work Ghosh and White(2001),using several updated star formation rate models,calculated the evolution of the X-ray luminosity of galaxies.They found that diﬀerent star formation models lead to very diﬀerent X-ray luminosity proﬁles,so the observed X-ray proﬁles can be used as probes of the star formation history.Finally they compared their models with theﬁrst Chandra deep imaging observations,to conclude that theseﬁrst results were consistent with current star formation models.Piro and Bildsten(2002)argued that the majority of LMXBs in theﬁeld of elliptical galaxies have red giant donors feeding a thermally unstable disk and stay in this transient phase for at least75%of their life.The very luminous X-ray sources (L x>1039erg s−1)detected in Chandra surveys have been suggested to be X-ray binaries with highly super-Eddington mass inﬂow near the accreting component.In elliptical galaxies these objects have been suggested by King2002to be micro-quasar-like,as these galaxies contain no high mass X-ray binaries(King2002).More recently Ivanova and Kalogera(2006) also argued that this bright end of the XLF is most likely dominated by transient LMXBs with BH accretors during outburst and it can be used to derive constraints on the BH mass function in LMXBs;they also showed that the standard assumption of a constant transient duty cycle(DC)across the whole population seems to be inconsistent with current observations.Semi-analytical population synthesis(PS)models of LMXBs have also been constructed for late type galaxies.Wu(2001)created a simple birth-death model,in which the lifetimes of the binaries are inversely proportional to their X-ray luminosity,and calculated the XLFs of spiral galaxies.His models reproduce some features,such as the luminosity break in the observed XLFs of spiral galaxies.The position of this break depends on the star formation history of the galaxy,and he suggested that it can be used as a probe of the galaxy’s merger history.The formation of LMXBs in GCs via dynamical interactions is less well studied,sinceapart from the binary stellar evolution,one has to also take into account the complex cluster dynamics.Bildsten and Deloye(2004)considered a semi-analytical model for accretion from degenerate donors onto NSs in ultracompact binaries and showed that binaries with orbital periods of8-10minutes and He or C/O white dwarf(WD)donors of0.06-0.08M⊙naturally provide the primary slope(-0.8for cumulative form)typically derived from XLFs of elliptical galaxies.Ultra-compact systems are predicted to form in the dense GC environment and have relatively short persistent lifetimes(<3×106yr)but they form continuously through dynamical interactions.Ivanova et al.(2007)presented PS studies of compact binaries con-taining NSs in dense GCs.They used StarTrack as their PS modeling tool in addition to a simpliﬁed treatment for the dynamical interactions.Their models produced a mixed popu-lation of LMXBs with red giant and MS donors,and ultra-compact X-ray binaries;relative formation rates can be comparable but the diﬀerent sub-populations have very diﬀerent lifetimes.In this paper we investigate the plausibility of an important contribution to the XLFs of these two galaxies from a primordial galacticﬁeld LMXB population using advanced PS simulations.In§2we describe brieﬂy the physics included in our PS code and explain in detail the way we are constructing the modeled XLFs and the treatment of transient LMXBs.We discuss the results of our simulations in§3:the modeled XLFs from diﬀerent models,a statistical comparison with the observed XLFs of the elliptical galaxies NGC3379 and NGC4278,and an analysis of the dependence of the modeled XLF properties on the PS parameters.Finally in§4we discuss the implication of ourﬁndings and the caveats of our methods.2.LMXB Population ModelsFor the models presented in this study we focus on LMXBs formed in the galactic ﬁeld as products of the evolution of isolated primordial binaries.The standard formation channel(Bhattacharya and van den Heuvel1991;Tauris and van den Heuvel2006)involves a primordial binary system with a large mass ratio;the more massive star evolves quickly to the giant branch and the system goes into a common envelope(CE)phase.During this phase,the less massive star,which is still dense and unevolved,orbits inside the envelope of the primary and is assumed to remain intact.The orbit of the system changes dramatically though,as orbital energy is lost due to friction between the the unevolved star and the envelope of the giant.Part of the lost orbital energy is used to expel the envelope of the giant star.The fraction of the lost orbital energy that is used to heat up the envelope of the giant star andﬁnally expel it,deﬁnes the CE eﬃciency factorαCE.The CE phaseresults in a binary system with an unevolved low mass main sequence(MS)star orbiting around the core of the massive star in a tighter orbit.The massive core soon reaches core collapse to form a compact object,either a NS or a BH and the binary orbit is altered due to mass loss and possible supernova kicks.If the binary does not get disrupted or merge in any of the stages described above,angular momentum loss mechanisms,such as magnetic braking,tides and gravitational wave radiation,will shrink further the orbit and the low-mass companion may evolve oﬀthe MS;The companion star eventually overﬂows its Roche lobe,transfering mass onto the compact object and initiating the system’s X-ray phase.An alternative formation channel for NS-LMXBs is through accretion induced collapse of a WD accretor into a NS.These systems have generally very low X-ray luminosity,and do not aﬀect the LMXB population in the luminosity range that we are interested in this paper.2.1.Synthesis Code:StarTrackWe perform the simulations presented here with StarTrack(Belczynski et al.2002, 2007),a advanced PS code that has been tested and calibrated using detailed mass trans-fer calculations and observations of binary populations,and incorporates all the important physical processes of binary evolution:(i)The evolution of single stars and non-interacting binary components,from ZAMS to remnant formation,is followed with analytic formulae (Hurley et al.2000).Various wind mass loss rates that vary with the stellar evolution-ary stage are incorporated into the code and their eﬀect on stellar evolution is taken into account.(ii)Throughout the course of binary evolution,the changes in all the orbital prop-erties are tracked.A set of four diﬀerential equations is numerically integrated,describing the evolution of orbital separation,eccentricity and component spins,which depend on tidal interactions as well as angular momentum losses associated with magnetic braking,gravi-tational radiation and stellar wind mass losses.(iii)All types of mass-transfer phases are calculated:stable driven by nuclear evolution or angular momentum loss and thermally or dynamically unstable.Any system entering the Roche lobe over-ﬂow(RLOF)is assumed to become immediately circularized and synchronized.If dynamical instability is encountered the binary may enter a CE phase.For the modeling of this phase we use the standard energy balance prescription.(iv)The SN explosion is treated taking into account mass-loss as well as SN asymmetries(through natal kicks to NSs and BHs at birth).The distribution of the SN kick magnitudes is inferred from observed velocities of radio pulsars.For this project we use the distribution derived by Hobbs et al.(2005)which is a single Maxwellian withσ=265km sec−1.It is however assumed that NS formation via electron capture or accretion induced collapse does not lead to SN kicks.(v)Finally,the X-ray luminosity of accreting binaries with NS and BH primaries(both for wind-fed and RLOF systems)is cal-culated.For RLOF-fed systems there is a distinction made between persistent and transient (systems that undergo thermal disk instability),while wind-fed systems are always consid-ered as persistent X-ray sources.The mass-transfer is conservative up to the Eddington limit for persistent X-ray binaries,while transients are allowed to have slightly super-Eddington luminosity(up to3×L Edd)(Taam et al.1997).In all cases we apply appropriate bolometric corrections(ηbol)to convert the bolometric luminosity to the observed Chandra band.A much more detailed description of all code elements,treatments of physical processes and implementation is provided in Belczynski et al.(2007).2.2.Model Parameters for NGC3379and NGC4278In this study we focus on trying to understand the XLF characteristics of the two ellip-tical galaxies NGC3379and NGC4278,observed with Chandra and reported by Kim et al. (2006a).In the development of our models we incorporate our current knowledge about the characteristics of the stellar population in these galaxies(see Table1).The observationally determined parameters of the stellar populations,such as their age and metallicity,or their total stellar mass,are similar for NGC3379and NGC4278.This allows us to develop the same models in our simulations for both of them.Terlevich and Forbes(2002)estimated the ages and metallicities of150elliptical and late type spiral galaxies using published hight quality spectral line indices.For NGC3379they are reporting an age of9.3Gyr and a metallicity of[Fe/H]=0.16,while for NGC4278the corresponding values are10.7Gyr and [Fe/H]=0.14.The two galaxies have very similar optical luminosity and assuming the same light to mass ratio,they should also have similar masses.Cappellari et al.(2006)used I-band observations from the Hubble Space Telescope to calculate the total stellar mass of the two galaxies and they found them to be8.6×1010M⊙and9.4×1010M⊙for N3379and N4278 respectively.The ratio of the integrated LMXB X-ray luminosity to the optical luminosity is 4times smaller for NGC3379which also has six times lower GC speciﬁc frequency compared to NGC4278(see Kim et al.2006a;Ashman and Zepf1998).Table1.Galaxy PropertiesParameter NGC3379NGC4278ReferencesThere is however a number of parameters in our models,for which we do not have any direct guidance from observations.We have no information about the star formation history of the two galaxies and thus we assume aδ-function like star formation episode at time t=0. Unknown are also the Initial mass function(IMF)and the distributions of orbital separation and eccentricity for the primordial binary systems.We adopt two diﬀerent initial mass functions:Scalo/Kroupa and Salpeter,while for the distributions of the orbital properties we follow the standard assumptions described in Belczynski et al.(2002).Other parameters that can aﬀect theﬁnal LMXB population are the binary fraction of the host galaxy,the magnetic braking law adopted and the CE eﬃciency(αCE)1.The speciﬁc parameters we used to model the ellipticals NGC3379and NGC4278are listed in Table2.2.3.Models for the X-ray Luminosity FunctionIn our models we keep track of all the binary properties,including the mass-transfer rate (˙M),as a function of time for populations of accreting NS and BH.We use the mass-transfer rates to identify the persistent and transient sources in our simulation results.Binaries for mass transfer rate higher than the critical rate˙M crit for the thermal disk instability (Dubus et al.1999;Menou et al.2002),are considered persistent sources and their X-ray luminosity(L x)is calculated directly from the mass transfer rate asGM a˙ML x=ηbolǫ1In our calculations,we combineαCE andλinto one CE parameter,whereλis a measure of the central concentration of the donor.In the rest of the text,whenever we mention the CE eﬃciencyαCE,we practically refer to the productαCE×λ(see Belczynski et al.(2007)for details.)Table2.Model Parameters for NGC3379and NGC4278Parameter Notation Valuewhere the radius of the accretor (R a )is 10km for a NS and 3Schwarzschild radii for a BH,ǫgives a conversion eﬃciency of gravitational binding energy to radiation associated with accretion onto a NS (surface accretion ǫ=1.0)and onto a BH (disk accretion ǫ=0.5),and ηbol is a factor that converts the bolometric luminosity to the X-ray luminosity in the Chandra energy band (0.3-8keV).For RLOF accreting BH this conversion factor is estimated to be ηbol =0.8(Miller et al.2001)while for RLOF accreting NS ηbol =0.55(Di Salvo et al.2002;Maccarone and Coppi 2003;Portegies Zwart et al.2004).The two correction factors ǫand ηbol are applied in both persistent sources and transient sources in outburst.In the context of the thermal disk instability model,mass transferring binaries with ˙M<˙M crit are considered transient sources,meaning that they spend most of their life in a quiescent state (T quiscent ),at which they are too faint to be detectable,and they occasionally go into an outburst.The fraction of the time that these systems are in outburst (T outburst )deﬁnes their DC:DC ≡T outburst 10h(2)We can generalize this relation to all transient LMXBs in galaxies other than our own,but we note that there has not been any observational work that shows that NS-LMXBs follow a similar trend.A more physical treatment is to assume that in the quiescent state the compact object does not accrete (or accretes an insigniﬁcant amount of mass)and matter from the donor is accumulated in the disk.In the outburst state all this matter is accreted onto the compact object emptying again the disk.Taking into account also that the X-ray luminosity probably cannot exceed L Edd by more than a factor of 2(cf.Taam et al.1997),we end up with adeﬁnition of the outburst luminosity as:L x =ηbol ǫ×min 3×L Edd ,GM a ˙Md DC (3)In the equation above,DC is unknown.Dobrotkaet al.(2006)studied accretion disk models for cataclysmic variables that are thought to experience the same thermal disk insta-bility (dwarf novae).They found a correlation between the DC of the system and the rate at which the donor star is losing mass ˙Md .The exact relation of these to quantities depends on the values of the disk’s viscosity parameters,but the general behavior can be approximated by:DC =˙M d R a ˙M d (5)The accretion disk models by Dobrotka et al.(2006)assume accretion onto a compact object with a hard surface and it is not obvious that the same results will apply for accre-tion onto a BH.In order to take into account all the available information,empirical and theoretical,about LMXB transient behavior,we treat BH and NS-LMXBs diﬀerently and deﬁne the outburst luminosity as:L x =ηbol ǫ× min 2×L Edd ,3×L EddP R a ˙M crit,for NS acc .(6)We note that for BH-LMXBs,we adopt a single DC value for simplicity and lack of other information,although we have no clear physical reason to believe that all BH systems have the same DC.It is believed that the DC of BH systems is smaller than that of NS systems and on the order of 5%(Tanaka and Shibazaki 1996).We found that in all our models NS accretors greatly outnumber BH accretors.BH systems only have an important contribution at high luminosities,where the error bars in the observed XLFs (see Kim et al.2006a)are too large to give us any tight constraints for our models.As we will see in section3.1,this treatment eq.(6)gives us the best agreement with observations.To construct the XLF we consider a snapshot of the whole population at the time we are interested in and we identify the LMXBs as transient or persistent.If a system is transient we decide whether it is in outburst or in quiescence according to its DC and either assign an outburst luminosity or discard the system as quiescent and hence too faint to contribute to the XLF.We then construct the XLF by calculating the cumulative X-ray luminosity distribution of the sources that are detectable(persistent and transient in outburst).We note that the age of the elliptical galaxies NGC3379and NGC4278and hence their LMXB population is known only to within∼1Gyr.Consequently we cannot just choose a unique snapshot of the population.Instead we construct the XLF by considering the time window of9-10Gyr and considering time slices separated by1Myr.We construct the XLF at each of these timeslices and we take the average to represent the XLF that corresponds to the time window of interest.Doing so,we also improve the statistics of our model sample.It is computationally impossible to evolve enough binaries that would correspond to the total initial number of binaries in an elliptical galaxy(∼109binaries).For each model we evolve106binaries which takes about two months of CPU time on a modern processor.We then normalize to the total mass of the galaxy in question,taking into account the initial binary fraction and the initial mass function.3.Results3.1.Exploring the parameter spaceOne of the implicit weaknesses of PS models is the large number of free parameters that one can vary andﬁne tune in order to get the desirable result.There are physical processes involved in the evolution of a binary system,such as stellar winds and magnetic braking,which are not fully understood.In this case various prescriptions are typically used to model them.Fortunately:(i)the result of interest to us in this study(XLF)is not sensitive to all of these model parameters;(ii)we can use some empirical knowledge from observations to constrain these parameters.We study a total of336models.In Table 3we list28combinations of PS input parameters we study(CE eﬃciency,initial mass function,wind strength)and in Table4we list the6diﬀerent prescriptions we use for the determination of the DC and outburst luminosity of transient LMXBs.For each of the parameter combination from Tables3and4,we try two prescriptions for the magnetic braking law,by Ivanova and Taam(2003)and Rappaport et al.(1983).Parameters not mentioned here are set as in the standard model considered in Belczynski et al.(2007).We name our models using a combination of a number from Table3which denotes the PS parameters of the models,a letter from Table4which denotes the prescription we use forTable3.Population Synthesis Models:For each of the models listed below we applied two magnetic braking law prescriptions.In the rest of the text the Ivanova and Taam(2003) prescription will be denoted with the“IT”superscript at the end of the model name,while the Rappaport et al.(1983)prescription with the“RVJ”superscript.ModelαCE IMFηwindTable4.Treatment of Transient LMXBs:We tried6diﬀerent prescriptions for the determination of the DC and outburst luminosity of transient LMXBs.Model Name L x,NS L x,BH DC NS DC BHthe treatment of transient LMXBs and a superscript which denotes the magnetic braking law used.The analysis of several population simulations enables us to identify general behaviors of how the LMXB population is aﬀected by changes to diﬀerent parameters.In Figure1we show the modeled XLFs for36selected models,separated in6panels.In each panels all the parameters that determine the formation and evolution of LMXBs are kept constant,and we just change the modeling of transient systems(their DC and outburst luminosity).The observed XLFs of the two galaxies,NGC3379and NGC4278,are also plotted for comparison.Weﬁnd that the stellar wind strength alters signiﬁcantly only the BH-LMXB population and thus only the high luminosity region of the modeled XLF.Stellar winds are important mainly in the evolution of massive stars which,depending on the wind prescription assumed, may lose a signiﬁcant part of their envelope.Weaker stellar winds result in more massive pre-supernova cores and thus to the formation of more BHs.On the other hand,the ini-tial mass function aﬀects the LMXB population globally.Aﬂatter(Salpeter)initial mass function increases the overall number of sources with luminosities above1037erg s−1but still gives results consistent with the observations.The modeled XLF is more sensitive to the CE eﬃciencyαCE.SmallerαCE values lead to more mergers among LMXB progeni-tors and therefore decrease the overall formation rate of LMXBs.If the CE eﬃciency is set to a very small value(αCE<0.2),it becomes very diﬃcult to form any LMXB at all (Kalogera and Webbink1998),while a high CE eﬃciencyαCE>0.6,clearly overproduce LMXBs.It is also clear that the adopted magnetic braking law greatly aﬀects the charac-teristics of the model LMXB population.The XLFs of all336models can be seen in the online material supplemental to this paper.The ratio of persistent to transient LMXBs produced from our models,including systems both in outburst and in quiescent,is on the order of1:20.So even if transient systems have a small DC(DC<10%),their contribution to the shape of the XLF is important and some times they may even dominate the population Piro and Bildsten(as discussed by2002). From this simple argument,we can understand that the shape of the modeled XLFs should be very sensitive to the DC and the outburst luminosity of transient systems.Figure1 conﬁrms this last statement,as we see that the larger variation in the shape of the modeled XLFs comes from diﬀerent prescriptions in the treatment of transient systems and not from changing the binary evolution parameters.。