CN1947062A - Method to determine the value of process parameters based on scatterometry data - Google Patents

Method to determine the value of process parameters based on scatterometry data Download PDF

Info

Publication number
CN1947062A
CN1947062A CN 200580012771 CN200580012771A CN1947062A CN 1947062 A CN1947062 A CN 1947062A CN 200580012771 CN200580012771 CN 200580012771 CN 200580012771 A CN200580012771 A CN 200580012771A CN 1947062 A CN1947062 A CN 1947062A
Authority
CN
China
Prior art keywords
collimating marks
group
substrate
calibration
measurement data
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN 200580012771
Other languages
Chinese (zh)
Inventor
H·A·J·克拉默
H·凡德拉恩
R·H·J·卡帕杰
A·G·M·基尔斯
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
ASML Netherlands BV
Original Assignee
ASML Netherlands BV
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by ASML Netherlands BV filed Critical ASML Netherlands BV
Publication of CN1947062A publication Critical patent/CN1947062A/en
Pending legal-status Critical Current

Links

Images

Landscapes

  • Exposure And Positioning Against Photoresist Photosensitive Materials (AREA)

Abstract

A method according to an embodiment includes obtaining calibration measurement data, with an optical detection apparatus, from a plurality of marker structure sets provided on a calibration substrate. Each marker structure set includes at least one calibration marker structure created using different known values of the process parameter. The method includes obtaining measurement data, with the optical detection apparatus, from at least on marker structure provided on a substrate and exposed using an unknown value of the process parameter; and determining the unknown value of the process parameter from the obtained measurement data by employing regression coefficients in a model based on the known values of the process parameter and the calibration measurement data.

Description

Determine the method for process parameter value based on the scatterometry data
The application requires the right of priority of the U.S. Provisional Patent Application sequence number 60/546,165 submitted on February 23rd, 2004, and the full content of this application is introduced in that this is for referencial use.
Technical field
The present invention relates to lithographic equipment and method.
Background technology
Lithographic equipment is that a kind of pattern with expectation is applied to the machine in the target portion of substrate.Lithographic equipment can be used in the making of integrated circuit (IC) for example.In this kind situation, can generate circuit pattern with picture construction as the mask corresponding to its single layer of IC, then such pattern (for example is imaged on substrate, silicon chip) the target portion on (for example, a part that comprises one or more circuit small pieces) on, substrate has radiation sensitive material layer (resist).Usually, single substrate all includes the grid of being made up of adjacent target portion, and these target portions will expose successively.Known lithographic equipment comprises usually said ledex and usually said scanner, and in usually said ledex, each target portion shines by the whole pattern that once exposes in this target portion; In usually said scanner, each target portion shines by using the projecting beam scan pattern along assigned direction (" scanning " direction), being parallel to simultaneously or being anti-parallel to this scanning direction substrate.
Although in this article, the application of concrete reference light engraving device when making IC should be appreciated that lithographic equipment described here can have other application.For example, it can be used for the manufacturing of integrated optics system, is used for the guiding of magnetic domain memory, LCD panel (LCD), thin-film head etc. and monitors pattern etc.It should be appreciated by those skilled in the art that, in this interchangeable range of application, herein any term " wafer " of Shi Yonging (wafer) or " circuit small pieces " (die) should be considered to respectively with more general term " substrate " (substrate) or " target portion " (targetportion) synonym.Herein the substrate of indication can be before exposure or after the exposure in for example guide rail (track) (being a kind of instrument that is commonly used to resist layer is applied on the substrate and the resist of exposure is developed), measure or checking tool in process.As long as be suitable for, herein openly can be applied to these and other substrate machining tool.In addition, can be more than once to substrate processing, for example in order to produce the IC of multilayer, thereby the term " substrate " that herein uses can also refer to the substrate of the layer that has comprised a plurality of processing.
Term used herein " radiation " and " beam " comprise all types of electromagnetic radiation, (for example comprise ultraviolet ray (UV) radiation, have 365,248,193,157 or the wavelength of 126nm) and far ultraviolet (EUV) radiation (for example having the wavelength in the 5-20nm scope), and the particle beams, as ion beam or electron beam.
Term used herein " picture construction " should be interpreted as referring to with extensively anticipating and can be used for giving patterned xsect to radiation beam (for example, projection fibers), thereby for example generates the structure of pattern in the target portion of substrate.Should be noted that the pattern of giving to beam may accord with to out of true the pattern of expecting in substrate target portion.Usually, give to the pattern of beam corresponding with the specific function layer of the device that in target portion, generates, as integrated circuit.
Picture construction can be transmission-type or reflection-type.The example of picture construction comprises mask, array of programmable mirrors and Programmable LCD plate.The notion of mask is known in photoetching, comprises phase shift type and various hybrid mask types as binary type, alternating phase-shift type and decay.An example of array of programmable mirrors is an arranged of utilizing tiny mirror, and each tiny mirror can tilt independently, so that along different directions reflection incident radiation beam; By this way, the beam to reflection carries out composition.
Support construction supports the weight of promptly bearing picture construction.Whether supporting construction for example remains in the vacuum environment as picture construction according to the orientation of picture construction, the design of lithographic equipment and other condition, is keeping picture construction.Support can be adopted mechanical grip, vacuum or other clamping technology, for example electrostatic clamp under the vacuum condition.Supporting construction can be for example framework or platform, can fix with the need or moves, and can guarantee that picture construction is on the desirable position, for example with respect to projection system.Here arbitrarily used term " marking-off plate " or " mask " can be considered to and more general term " picture construction " synonym.
Term used herein " projection system " should broadly be interpreted as containing various types of projection systems, comprise dioptric system, reflective optics and reflection and refraction optical system, as long as they are suitable for for example used exposing radiation, the factor that perhaps is suitable for other is such as the immersion liquid of using or the vacuum of use.Here arbitrarily used term " camera lens " can be considered to and more general term " projection system " synonym.
Illuminator also can contain various types of opticses, comprise that configuration is used for guiding, be shaped or controls refraction, reflection and the catadioptric optical parts of the projection fibers of radiation, and these parts also can be called " camera lens " below jointly or separately.
Lithographic equipment can be the type with two substrate platforms (twin-stage) or more substrate platforms (and/or two or more mask platform).In this " multistage " machine, can walk abreast and use these additional stations, perhaps can on one or more platform, carry out preparation process, and one or more other be used for exposure.
Lithographic equipment can also be this type, and promptly wherein substrate is immersed in the liquid for example in the water with relative high index, with last element that fills up projection system and the space between the substrate.Immersion liquid can also be applied to other spaces in the lithographic equipment, for example, and the space between first element of mask and projection system.Can adopt immersion technique to improve the numerical aperture of projection system.
At present, the continuous development trend towards the device density of littler design feature and Geng Gao needs more high-resolution photoetching technique.In order to satisfy these needs, expectation can be controlled photoetching process as much as possible in detail.Needing wherein two most important technological parameters of precise monitoring and control is dosage (dose) and focal length (focus).Usually, monitor and control these parameters by measuring critical dimension (critical dimension) variation (variation) (CD).Yet, when measuring the CD variation, be difficult to distinguish dosage and focal length data.
Usually, use special-purpose member or a plurality of parts and special-purpose weights and measures or time-consuming weights and measures combined.Focal length for example can be determined with phase shift focal length monitor.Can cause aliasing error when error appears in focal length, and aliasing error can detect at an easy rate with the overlapping instrument of reading.In second kind of technology, realize the monitoring of focusing by the notion of using line end to shorten.Yet when using this technology, the sign that defocuses is very difficult to determine.In addition, most of current techniques only is applicable to test structure.
Therefore, quality to the pattern of lithographic equipment exposure is monitored the technology that just needs a kind of fast and stable, this technology can be used in a lot of positions on all types of substrates (as test wafer or product wafer) that will expose, for example on chip area inside or the groove.A kind of light quantity technology that is called scatterometry can satisfy these requirements to a certain extent.Term used herein " optics " and " light " are contained all types of electromagnetic radiation, comprise wavelength be 400-1500nm light, ultraviolet (UV) radiation (for example, have 365,248,193,157 or the wavelength of 126nm) and far ultraviolet (EUV) radiation (for example, has the wavelength in the 5-20nm scope), and the particle beams, as ion beam or electron beam.
In scatterometry, light beam is guided towards a target, and normally a kind of custom-designed structure of this target is as diffraction grating.Then, target reflection, refraction and/or diffraction light.Finally, the detector that can be contained appropriate sensor from the light of target detects.The detection of detector can be reflection or transmission, in order to measure diffraction light and/or non-diffracted light.For incident light, promptly be directed into the light at target place, can change one or more groups character simultaneously.Term used herein " scatterometry " and " scatterometer " contain can produce light and light and target interaction post analysis light, all types of measuring techniques and survey instrument.Term " scatterometer " comprises for example ellipsometer and scanning electron microscope (SEM).Term used herein " spectrum " is encompassed in all types of forms that can be detected after light and the target interaction.Therefore, it comprises the image that is generated by the scattered electron among the SEM.
Scatterometry is used to determine the value of technological parameter traditionally, as focal length and dosage.Yet,, can make several hypothesis usually to the relation between technological parameter and the scatterometry parameter.The example of the relation of these hypothesis has the linear relationship between focal length and side wall angle (slope of linear structure side), and the linear relationship between dosage and half CD (the linear structure width that half is highly located at it).In fact, single scatterometry parameter and technological parameter may not be unique as the relation between focal length or the dosage.May be except that focal length, for example the character at other effect oppose side wall angle also has contribution.According to above-mentioned hypothesis, these effects will be interpreted as focal length by absurd explaining by oneself.
The spectrum that detects (perhaps, using under the situation of the particle beams, the signal that detects can be image rather than spectrum) by be stored in data in the database and compare and analyzed.What is called between the spectrum in spectrum that detects and database " optimum matching " has determined to describe best the parameter value of target structure.For the purpose of photoetching, the parameter value that identifies, promptly focal length and dosage can be applied to improving the performance of lithographic equipment.The quality of control of photoetching process parameter and monitoring is depended on significantly the quality of database.The theoretical spectral that makes up by for different scatterometry calculation of parameter values is housed in the database usually, these scatterometry parameters for example have grating parameter as grating height, live width and side wall angle, different substrate parameters, for example material character reaches the layer relevant character interior with the previous substrate of handling.Understand easily, it is very consuming time and very complicated to generate extremely reliable database, especially when the character of the substrate that will expose often changes.
In addition, the scatterometry parameter as the optical constant of the thickness and the material therefor of underlying bed, all is extremely to be difficult to definitely in making situation.For this reason, the use experience data have been advised, i.e. the data that test obtains.(for example, referring to people's such as Allgair Yield Management Solutions, Summer 2002, pp8-13).In this example, experience database is to be produced by a substrate with multiple structure, and this substrate is handled with containing technological parameter processing space (process space), one group of variation that will control.Yet, as mentioning in this list of references, because to the required controlling level of technological parameter and by " variation naturally ", the material impact of the noise that promptly involuntary variation of inducing causes, the characteristic of these structures is not to be footy.
Summary of the invention
Embodiment of the present invention comprise and are used for determining at least one and the method for the relevant technological parameter of photoetching method that utilizes empirical data.A kind of embodiment provides a kind of method that is used for determining at least one technological parameter, and this method comprises:
Obtain the calibration measurement data a plurality of collimating marks structural group on being arranged on calibration target, in described a plurality of collimating marks structural group each all comprises at least one collimating marks structure, and the collimating marks structure in the different collimating marks structural group adopts the different given values of described at least one technological parameter to generate;
Described given value by using described at least one technological parameter is also utilized regression technique to described calibration measurement data, determines mathematical model, and described mathematical model comprises several regression coefficients;
Obtain measured value from least one mark structure that is arranged on the target, described at least one mark structure is to use the unknown-value of described at least one technological parameter to make; And
By utilizing the described regression coefficient of described mathematical model, be the unknown-value that described target is determined described at least one technological parameter from the described measurement data that obtains.
In another embodiment of the invention, a kind of system that is used for determining at least one technological parameter is provided, this system comprises:
Detector, layout is used for obtaining the calibration measurement data a plurality of collimating marks structural group on being arranged on calibration target, in described a plurality of collimating marks structural group each all comprises at least one collimating marks structure, and the collimating marks structure in the different collimating marks structural group is to adopt the different given values of described at least one technological parameter to generate;
Processor unit is stored a mathematical model, and this mathematical model is to use regression technique and definite by the described given value of using described at least one technological parameter and to described calibration measurement data, and described mathematical model comprises several regression coefficients;
Described processor unit is arranged and is used for obtaining measurement data from least one mark structure that is arranged on the target, described at least one mark structure uses the unknown-value of described at least one technological parameter to make, and be used for from the described measurement data that obtains, determining this unknown-value of described at least one technological parameter for described target by utilizing the described regression coefficient of described mathematical model.
In one embodiment of the invention, this system comprises lithographic equipment, and this lithographic equipment comprises: illuminator, and configuration is used to provide the projection fibers of radiation; Supporting construction, configuration is used for supporting picture construction, and this picture construction is used for giving pattern to radiation beam in xsect; The substrate platform, configuration is used for laying substrate; And projection system, configuration is used for the light beam of patterning is projected in the target portion of substrate.
In one embodiment of the invention, provide a kind of semiconductor devices of making according to the method for any embodiment of the present invention disclosed herein of using.
In one embodiment, this system comprises lithographic equipment, and this lithographic equipment comprises: illuminator, and configuration is used to provide radiation beam; Supporting construction, configuration is used for supporting picture construction, and this picture construction is used for giving pattern to radiation beam in xsect; The substrate platform, configuration is used for laying the substrate with at least one mark structure; And projection system, configuration is used for the light beam of patterning is projected in the target portion of substrate.
The invention still further relates to a kind of semiconductor devices of using according to system's manufacturing of any previous embodiments.
Description of drawings
Referring now to appended schematic figures, describe each embodiment of the present invention, described embodiment is as just example, and corresponding in the accompanying drawings reference symbol is represented corresponding parts, in the accompanying drawings:
Fig. 1 has drawn a kind of lithographic equipment according to one embodiment of the invention;
Fig. 2 has drawn the scatterometer of prior art;
Fig. 3 has drawn the functional sequence based on the method for database;
Fig. 4 drawn use the calibration spectrum that measures, based on the functional sequence of the method for database;
Fig. 5 a, 5b show according to one embodiment of the invention, the functional block diagram in two stages of expression;
Fig. 6 has drawn according to one embodiment of the invention, and the functional block diagram of regression concept is described;
Fig. 7 a, 7b illustrate according to one embodiment of the invention, with the notion of harmonic wave division and the notion of dividing with principal component;
Fig. 8 a, 8b illustrate according to one embodiment of the invention, the top view of the various combination of mark structure;
Fig. 9 has drawn the etching system according to one embodiment of the invention.
Embodiment
Fig. 1 schematically represents a kind of according to lithographic equipment of the present invention.This device comprises: illuminator (luminaire) IL, and configuration is used to provide the projection fibers PB of radiation (for example, the radiation of UV radiation or other wavelength); First supporting construction (for example, mask platform) MT, configuration is used for supporting picture construction (for example, mask) MA and is connected to the first positioning equipment PM, and this first positioning equipment PM configuration is used for accurately locating picture construction with respect to projection system PL.This device (for example also comprises the substrate platform, wafer station) WT, configuration (for example is used for laying substrate, apply the wafer of resist) W and be connected to the second positioning equipment PW, this second positioning equipment PW configuration is used for accurately locating substrate with respect to projection system (camera lens) PL, projection system (for example, refractive lens system) PL (camera lens) configuration is used for being imaged on the C of target portion (for example, comprising one or more circuit small pieces) of substrate W being given by picture construction MA to the pattern of projection fibers PB.
As shown here, this device is transmission-type (for example, utilizing transmission mask).Alternatively, this device can be reflection-type (for example, utilizing the above-mentioned array of programmable mirrors type of mentioning).
Luminaire IL receives the radiation beam that comes from radiation source S O.For example when radiation source was excimer laser, this radiation source and lithographic equipment can be entities separately.In these cases, radiation source is not considered to form the part of lithographic equipment, radiation beam sends from radiation source S O, passes to luminaire IL under the help of beam delivery system BD, and this beam delivery system BD comprises for example suitable directing mirror and/or beam expander.In other situation, for example when radiation source was mercury lamp, radiation source can be the device part of the whole.Radiation source S O and luminaire IL can be known as radiating system with beam delivery system BD (if desired).
Luminaire IL can comprise conditioning equipment AM, and this conditioning equipment AM is configured to regulate the angle intensity distributions of beam.Usually, in the pupil plane of irradiator at least the external diameter scope of intensity distributions and/or inside diameter ranges (be called respectively usually σ-outside and σ-Nei) can regulate.In addition, luminaire IL generally also comprises various miscellaneous parts, for example integrator IN and condenser CO.Luminaire can be provided at have in its xsect desirable uniformity coefficient and intensity distributions, through adjusting the radiation beam handled, be called projection fibers PB.
Projection fibers PB incides on the mask MA, and mask MA is fixed on the mask table MT.After passing mask MA, projection fibers PB passes camera lens PL, and camera lens PL focuses on projection fibers on the C of target portion of substrate W.(for example, interferometry equipment under) the help, substrate platform WT can accurately move, thereby for example the different C of target portion is positioned in the path of beam PB at the second positioning equipment PW and position transducer IF.Similarly, the first positioning equipment PM and another position transducer (not illustrating clearly in Fig. 1) can be used for respect to the path of beam PB location mask MA accurately, for example at P9 after mask data storehouse machinery obtains, perhaps in scan period.Usually, moving of Target Station MT and WT can be achieved under the help of long stroke module (coarse localization) and short stroke module (accurately location), and long stroke module and short stroke module can form the part of positioning equipment PM and PW.Yet in the situation of ledex (relative with scanner), mask table MT can only be connected with the short stroke actuator, perhaps can fix.Mask MA and substrate W can use mask alignment mark M1, M2 and base plate alignment mark P1, P2 to aim at.
Shown device can use according to following preferred mode:
1. in step mode, it is motionless substantially that mask table MT and substrate platform WT keep, and projected on the C of target portion by (that is single static exposure) quickly and give to the whole pattern of projection fibers.Then, along X and/or Y direction moving substrate platform WT, the feasible C of target portion that can expose different.In step mode, the full-size of exposure field defines the size of the C of target portion of imaging in single static exposure.
2. in scan pattern, to mask table MT and substrate platform WT synchronous scanning, the pattern that will give simultaneously to projection fibers projects the C of target portion upward (that is single dynamic exposure).Amplification (dwindling) multiple and image inversion character that speed that substrate platform WT moves with respect to mask table MT and direction depend on projection system PL.In scan pattern, the full-size of exposure field defines at the hit width (along non-direction of scanning) of portion of single dynamic exposure, and the length that scanning is moved has been determined the height (along the direction of scanning) of target portion.
3. in another kind of pattern, it is motionless substantially that mask table MT keeps, and laying picture construction able to programme, and substrate platform WT is moved or is scanned, and the pattern that will give simultaneously to projection fibers projects on the C of target portion.In this pattern, the general using impulse radiation source, and after each the moving of substrate platform WT or scan period between continuous radiation, as required picture construction able to programme is upgraded.This operator scheme is easy to be applied in and adopts patterning device able to programme, in the maskless photoetching as array of programmable mirrors above-mentioned.
Can also adopt the combination or the variation of above-mentioned use pattern, or diverse use pattern.
Fig. 2 represents the scatterometer of prior art.This scatterometer comprises light source 1, the structure 5 (being generally certain type grating) on the substrate W that light source 1 lead beam 2 directives will expose, and substrate W is positioned on the substrate platform WT.This scatterometer also comprises detector 4.Detector 4 is connected to (little) processor 9, and microprocessor 9 is connected to storer 10.Reflection and/or diffraction take place at appropriate configuration 5 places that are positioned on the substrate W surface in light beam 2.The spectrum of folded light beam is detected device 4 and surveys.Light beam 2 can be directed directive substrate W at a certain angle, as shown in Figure 2, but also can be directed perpendicular to substrate W.The notion of scatterometry has multiple, and one or more groups character that wherein is directed into the light of appropriate configuration can change simultaneously.The example of one group of character has one group of wavelength, one group of incident angle, one group of polarization state or one group of phase place and/or phase differential.Detector can be provided for surveying the combination of aforesaid group or aforementioned group, and detector can comprise the different piece of one or more sensors with record reflected light and/or diffraction light.
Fig. 3 is illustrated in the functional overview that adopts in the scatterometer based on the method for database.This database generally can be by being made up for different scatterometry calculation of parameter spectrum, these scatterometry parameters for example are structural parameters, the live width of each line, line height, side wall angle etc. in the picture structure 5, the thickness of each underlying bed (not having composition) below these lines, and with the optical constant of light beam 2 interactional all materials.Before actual physical structure 5 is measured, may need to define the aforementioned parameters relevant with the structure 5 of this kind particular type.For each preset range in the parameter of these definition,, calculate the spectrum of the light of being modulated by structure 5 with processor 9, and spectrum is stored in the spectra database of storer 10 in operation 301.
Then, can understand ground, can carry out Theoretical Calculation known structure as those of ordinary skill in the art.For instance, when database has been filled enough spectrum by processor 9, when these spectrum are enough to contain the desired extent of spectrum property of the practical structures 5 that will measure, practical structures 5 is measured.Subsequently, this method proceeds to operation 302, and wherein processor 9 compares the spectrum of the measure spectrum of practical structures 5 and a plurality of storages in the spectra database of storer 10.Perhaps, also can use real-time coupling.
Then, by adopting interpolation algorithm, in operation 303, extract " optimum matching ", and identify the parameter value that is used to produce the parameter of the spectrum that extracts corresponding to those from storer 10.For instance, when the spectrum that makes up as the spectrum that measures and the value B3 of value A1 by adopting parameter A and B parameter was the most similar, processor 9 finally provided output { A1, B3}.
Can adopt strict diffraction modeling algorithm, calculate spectrum in the spectra database as rigorous couple-wave analysis method (RCWA).This complicated algorithm that is used to calculate at the spectrum of the spectra database stored of storer 10 in other respects, may need to understand in advance the optical property of material therefor.In fact, particularly for the product wafer, the value of having only some character in these character is known, therefore generally can adopt approximate.And in making situation, the character of different structure is not fully known in the underlying bed.Therefore, the conventional method based on database may be very complicated, and this has limited the daily use in making situation.
In the following description, with reference dose and focal length as exemplary technological parameter.Yet, should be appreciated that when using other photoetching process parameter, can use embodiment of the present invention in a similar fashion.Other examples of the technological parameter that can adopt for example comprise the guide rail parameter relevant with dosage, the variation of live width on the marking-off plate, and the variation of marking-off plate and marking-off plate, the aberration of projection camera lens throws glittering of camera lens, and the angle of the light of illumination marking-off plate distributes.
Fig. 4 represents according to one embodiment of the invention, based on the functional representation of the scatter measurement method of database.In this method, directly use the calibration spectrum that measures, rather than theoretical spectroscopic data, comparing with the spectrum that on actual physical structure, measures, this actual physical structure can for example be a diffraction structure, such as grating.Before to the actual physical structure measurement, on calibration substrate, calibrate.
In one embodiment of the invention, calibration substrate is provided with a plurality of calibration structures, and the shape that each calibration structure has can be compared with the physical arrangement that will measure substantially.Each calibration structure can have unique position on calibration substrate, and utilizes technological parameter to construct as the combination of the unique value of focal length and exposure (dosage).In one embodiment of the invention, the value of first technological parameter changes along first direction across substrate, and second technological parameter can change along the second direction that is basically perpendicular to first direction.In one embodiment of the invention, first is focal length and dosage with second technological parameter.In this case, calibration substrate is called focal length exposure matrix (FEM).In the following description, will explain the design of each embodiment of the present invention with reference to FEM.Yet, be appreciated that in other embodiments of the present invention and also can adopt alternative matrix.
In one embodiment of the invention, this method starts from operation 401, wherein measures calibration spectrum with FEM, then calibration spectrum is stored in the storer 10 together with the information relevant with the value of focal length that is used to make this spectrum and dosage.Then, measure the spectrum that incides the structural light of actual physical.Then, in operation 402, this spectrum that measures is compared with the spectrum in being stored in storer 10.Then this method proceeds to operation 403, wherein extracts " optimum matching " from storer 10.In this stage, obtain the value of dosage and focal length from the spectrum that extracts.For instance, in Fig. 4, being confirmed as in the described spectrum that measures and " optimum matching " between the spectrum that measures on the structure that is provided by FEM is spectrum corresponding to the value E2 of the value F2 of focal length and exposure (dosage).
A potential advantage that is appreciated that at least some embodiments as shown in Figure 4 is in order to determine that described parameter does not need to understand in advance the optical property of material.
Yet, as describe in the above any based in the method for database like that, be discretize to the determined value of technological parameter of selection.In addition, by " natural variation ", the noise that promptly involuntary variation of inducing is introduced may produce remarkable influence to the identification of the value of the technological parameter selected in calibrating operation.It is desirable to and to reduce to minimum by the interference that this natural variation causes identification.
The source that natural variation produces can be as follows.In scanner, natural variation may be relevant with the error of focal length and exposure dose at random, has unique Jiao Ju ﹠amp for each; The single exposure of dosage setting, the error of focal length and exposure dose and inequality at random.In guide rail, natural variation may relevant with uneven processing on the wafer (this and dosage be that part is relevant).In wafer, natural variation may be relevant with uneven underlying bed on the wafer.In scatterometer, natural variation may be with heat, machinery be relevant with electrical noise.
Fig. 5 a, 5b illustrate the functional block diagram of one embodiment of the invention.In this embodiment, by utilize regression technique (regression technique) at calibration phase, use calibration spectrum to form a mathematical model.Then, the operational phase by the mathematical model that utilization obtains, draw the technological parameter that is used for making the practical structures of measuring.Fig. 5 a is illustrated in such embodiment of the present invention, in the used method of calibration phase.This method starts from operation 501, wherein measures calibration spectrum with a plurality of calibration structures, and calibration spectrum is stored in the storer 10.These calibration structures are made up with one group of known technological parameter, and for each calibration structure, these technological parameters all are different.For example, when these technological parameters were focal length and dosage, this method was at first measured calibration structure with FEM, and the spectrum that measures is stored in the storer 10.
Then, this method proceeds to operation 502, wherein with the processor that is connected to storer 10 calibration spectrum of storage is carried out regretional analysis.Sort processor can be processor 9 in one embodiment of the invention, in other embodiments of the present invention, can also be different processors.Then, this method proceeds to operation 503, wherein determines to be stored in the mathematical model in the storer.This mathematical model has defined calibration spectrum and has been used to make relation between the technological parameter of the calibration structure of measuring this calibration spectrum thereon.Sort memory can be the storer 10 that is connected to processor in one embodiment of the invention, but also can be the different storer that is connected to processor in other embodiment of the present invention.
Fig. 5 b illustrates the method according to one embodiment of the invention, and this method can be carried out by processor 9, and it uses the model of described acquisition to draw the value of the technological parameter of selection according to the measurement of carrying out on " reality " structure on the substrate.This method starts from operation 511, wherein to " reality " structure measurement response signal on the substrate.The signal that measures can be a spectrum, as the input of model.Then, this method proceeds to operation 512, wherein determines the expectation value of selected technological parameter.Then, this method proceeds to operation 513, wherein in photoetching process or manually or automatically use determined technological parameter, to proofread and correct for example external setting-up of lithographic equipment, setting (for example, substrate platform WT's moves) etc. is set, located to picture dosage setting, focal length.
Being appreciated that can reduce to the effect of natural variation minimum in each embodiment of the present invention.Because the natural variation of the technological parameter of selecting is included in the calibration process, so the model that generates can be independent of the natural variation of this technological parameter.Minimum in order better the effect of natural variation to be reduced to, it is desirable to use random variation (for example, calibration chip can be made and use random variation).And, if the natural variation of technological parameter is known, so can be when calibration phase forms model with this natural variation as independent input.Here, " independent " input refers to other input, perhaps refers to this input and can substitute the process deviation that those are deliberately induced.
In one embodiment of the invention, used regression technique can be linear or nonlinear in the Return Law.In one embodiment of the invention, can also use neural network.These technology can be applied to provide interpolation between the calibration point of model, and/or reduce noise.
Fig. 6 represents the functional block diagram according to the regression technique of another embodiment of the invention.This design based on be iterative process, wherein use response signal X and the one group of Prediction Parameters Y measure to calculate regression coefficient b, form mathematical model thus, regression coefficient b combines X and Y.Prediction Parameters Y be with check in the relevant parameter of technological parameter.This method starts from operation 601, and one group of Prediction Parameters Y wherein is provided, and this method proceeds to operation 603 then, wherein measures response of structure signal X on the substrate.Prediction Parameters Y and the response signal X that measures are used as the input of mathematical model, in operation 605, the computation model change regression coefficient b.Then, in operation 607, verify the importance (significance) of all regression coefficient b.This control operation determines whether this mathematical model is healthy and strong.In operation 609, from mathematical model, remove the unimportant regression coefficient of meaning, and repeat to return with the regression coefficient of decreased number.Repeat operation 605 and 607, regression coefficients all in mathematical model is all very important always.Next, this method proceeds to operation 611, wherein uses regression results to come to determine Prediction Parameters Y for new response signal X.
In one embodiment of the invention, can adopt linear regression (MLR) that transformation of data is become information.When response signal seldom the time, there are some suitable situations, be also referred to as the factor sometimes.Significantly redundant in these factors, just under the situation of conllinear, perhaps when they and Prediction Parameters Y had concerning of fine understanding, MLR was very useful.Yet if any one in these three conditions do not satisfy, MLR is invalid or unfavorable so.Embodiment of the present invention have comprised based on existing one or more such conditions to use those methods of MLR.
In one embodiment of the invention, use the spectrum that is measured by scatterometer, estimation photoetching process parameter is such as the value of dosage and focal length.Usually, the described factor comprises spectrum hundreds of and the height conllinear.In this case, Prediction Parameters Y is the value of photoetching process parameter.
Fig. 7 a, 7b illustrate each example of the decomposition technique that can be used in the embodiment of the present invention.Adopt Fourier analysis in first kind of technology shown in Fig. 7 a, this analysis based on principle be that signal can be described with the summation of primary harmonic function, about each function rises with specific weight factor.For example, signal S1, S2, S3 and the S4 among Fig. 7 a can be described to respectively H1 and H2 with weight factor [1 ,-1], [1 ,-1/2], [1 ,+1/2] and [1,1] effect with.
Second kind of technology is a kind of similar techniques, its based on principle be that signal can be described with the summation of some principal components, each principal component works with specific weight factor.The number of principal component can change.Fig. 7 b illustrates four exemplary scatterometry spectrum (F1, F2, F3 and F4), and these four spectrum can be described by using weight factor [1 ,-1], [1 ,-1/2], [1 ,+1/2] and [1,1] combination principal component PC1 and PC2 respectively.In an embodiment that is used for regretional analysis of the present invention, can utilize be similar to above-mentioned mention but be not limited to the decomposition technique of these examples.For example, return in the situation of (PCR) in principal component, the principal component that extracts from the response signal X that measures can substitute the X factor as shown in Figure 6, as the input of mathematical model.
Except that the decomposition technique of these two kinds of descriptions, can also adopt other decomposition technique in embodiments of the invention.The example of these technology comprises the decomposition technique based on local least square method (PLS) modelling and the modeled notion of non-linear PLS, for example as people such as Wold at Chemometrics and IntelligentLaboratory Systems, as described in 7 (1989) 53-65.
In one embodiment of the invention, before presenting spectroscopic data to model, can carry out certain type pre-service.Pre-service can improve the result of model.The example that is applicable to pretreatment operation of the present invention deducts mean value, and divided by standard deviation, weighting or selective scattering measurand are such as angle, wavelength and polarization state.As a result, before feeds of data is given model, just can remove the data at some wavelength place.
In each embodiment of the present invention, in calibration process and measuring process, can adopt the mark structure of a more than class.Thereby the present invention is applicable to a plurality of calibration structure groups, and wherein every group comprises one or more (different) calibration structure, and the number of the calibration structure in each calibration structure group can be different.In addition, in these calibration structure groups and/or in twos between the calibration structure group, calibration structure can be different.It is desirable to, in every group, dissimilar mark structures is closely located on calibration substrate each other with closing on.And, wish that also calibration measurement and sample measurement are essentially identical (for example, identical pre-service, identical mark or marker combination and/or identical chip-type) at least in some respects.Before being used, can add (append) at resulting spectrum on these marks each other by mathematical model.But, in one embodiment of the invention,, also these spectral combination can be got up, thereby obtain a kind of combination " spectrum " that can be used by model by means of certain type mathematical operation.
Fig. 8 a, 8b illustrate can be with the top view of the calibration substrate 801 of in embodiments of the invention the mark structure that is provided with combination.In Fig. 8 a, first mark structure 802 is included in several patterns that does not form on the top of patterned layer.Second mark structure 803 does not comprise described pattern, and is only formed by patterned layer not.In Fig. 8 a, only show a group with first and second mark structures 802,803.But,,, on identical calibration substrate 801, produce several such groups to not adopting different technical parameters on the same group in order to implement calibration steps of the present invention.In scatterometry, 803 of second mark structures reflect the variation in the patterned layer not, and the pattern of first mark structure 802 does not increase its contribution in the contribution of patterned layer at these.Cut down the not contribution of patterned layer among the scatterometry result that can be used on first mark structure 802, obtain in the scatterometry result who obtains on second mark structure 803.The example that can implement this reduction operation comprises the spectrum that deducts second mark structure 803 from the spectrum of first mark structure 802, the spectrum of second mark structure 803 and the spectrum of first mark structure 802 are consistent, then with the input of remainder as mathematical model.
In Fig. 8 b, show according to one embodiment of the invention the various combination of the group that constitutes by two mark structures on the substrate 801.Though shown in Fig. 8 b is a group,, can on substrate, produce a plurality of such groups to not adopting different technological parameters on the same group in order to implement method of the present invention.First mark structure 802 for example comprises the pattern identical with first mark structure 802 of Fig. 8 a.Yet, different with second mark structure 803 of Fig. 8 a, be patterned at second mark structure 804 shown in Fig. 8 b.In this embodiment of the present invention, mark structure 802,804 all is patterned, but the pattern of each mark structure is different.Because these mark structures have different susceptibilitys to the photoetching process parameter, so just can separate these technological parameters better.Be appreciated that in other embodiments of the present invention, can also utilize other combinations of patterning mark structure.In one embodiment of the invention, can use more than two mark structures.
When focal length is when adopting one of technological parameter that an embodiment disclosed herein measures, by utilizing down the further optimization of one of surface technology.In one embodiment of the invention, in order to produce the bigger variation of the spectral shape that every nanometer defocuses, can use littler mark structure, this is because these structures have the less depth of focus.In another embodiment of the invention, in order to improve the susceptibility that focusing changes, can use structure with more sidewalls, for example, use half contact hole of isolating or 1 point of isolating to replace using line.In another embodiment of the invention, can also use a kind of resist that under defocusing, can show bigger spectrum change.Yet in making situation, this is selected may be inapplicable.
Should be appreciated that the substrate of any kind, for example product wafer or testing wafer can use in the application of embodiment of the present invention.It is also understood that the actual physical structure that will measure according to the needs of using, can be placed on any position on the substrate, for example, it is inner or be placed on the groove to be placed on chip area.In addition, in some embodiments of the present invention, the luminous point of scatterometer can be the same with chip area or exposure field big.The luminous point of this size can be determined the skew and the exposure field of each chip respectively apace.
That may wish arranges the photoetching process parameter across FEM in slippage (shuffled) mode, as the value of focal length and dosage.In addition, technological parameter can also increase from a side direction opposite side of substrate.Thereby the technology variation meeting that is caused by the source of lithographic equipment outside produces remarkable influence to calibration result, and these technologies variations are linear across substrate usually, and/or is rotational symmetric with respect to the center of substrate.For example, by these are worth across the FEM slippage, the technological parameter that induce these outsides can be eliminated significantly.
In one embodiment of the invention, in order to limit the purpose of (qualification), the technology variation that can adopt little FEM to eliminate to induce the outside.The very little part of this FEM covered substrate.The technological parameter of therefore, can case of external inducing can be ignored.
In manufacture process, handle many identical wafers usually one by one.Set for the best of concrete photolithographic fabrication process in case determine lithographic equipment, these settings must be kept in tight control limit so.In one embodiment of the invention, these are set with automated session control (APC) and keep.In this case, will be to the measurement of product wafer code fo practice, thus allow to carry out FEEDBACK CONTROL.
The present invention can be respectively applied for guide rail and lithographic equipment.The present invention can be used to rotate the knob (knob) on (twist) guide rail or the lithographic equipment, and does not directly relate to the technological parameter that will control.By measuring relevant technological parameter subsequently and utilizing the present invention, can obtain the effect that this knob rotates, and can select the best of knob to set.With compare in the past, can avoid adopting the information that scanning electron microscopy (SEM) and these technology employing off line measurements of electricity wire width measuring (ELM) obtain expecting that is similar to.
Fig. 9 represents the etching system according to one embodiment of the invention.In this embodiment, will be to scatterometer 902 with the substrate transfer printing (in the back of developing with guide rail) of lithographic equipment 901 exposure.Scatterometer 902 is connected on the control module 903 that comprises processor 9 and storer 10.Lithographic equipment 901 prints out the mark structure that is suitable for scatterometry by using technological parameter-focal length and preestablishing of dosage, at first generates FEM.Then, substrate is transmitted (910) to scatterometer 902.Scatterometer 902 is measured calibration spectrum and the spectrum that measures is stored (911) in the calibration data base 904 of storer 10.
Then, the mark structure that lithographic equipment 901 usefulness are identical carries out composition to the product substrate.Then, substrate is transmitted (912) to scatterometer 902.Scatterometer 902 is measured from the spectrum of the light of mark structure reflection, and this mark structure is produced by lithographic equipment 901.Then, spectrum is fed to the mathematical model 905 that (913) can be used by processor 9.This mathematical model 905 is used by processor 9, so that the calibration spectrum that will be stored in the calibration data base 904 compares with the spectrum of measuring on mark structure, subsequently, processor 9 draws the value of the parameter that will control by utilizing regression technique, as dosage and focal length.
At last, the value of processor 9 these parameters that will draw offers lithographic equipment 901.Lithographic equipment 901 can adopt for example these value that draws monitoring skew with respect to reference state in device.Subsequently, in feedback signal, use the described value that draws to proofread and correct these skews.In this case, lithographic equipment 901 is provided with the correction control module, and this correction control module uses the correction signal that applies that skew is compensated.This correction control module 903 can be configured to the height of control example such as substrate platform WT, to improve focal length.
In an alternate embodiment of the present invention, the value of drawing with these parameters does not offer lithographic equipment 901, but offers other entity, as guide rail, terminal or display.In the latter's situation, the operator who is responsible for operational light engraving device 901 for example can verify, and whether these values of drawing fall in the control limit.In another embodiment of the invention, mathematical model 905 and/or calibration data base 904 can be arranged in the entity different with control module 903.In one embodiment of the invention, lithographic equipment 901 can be connected on the identical guide rail with scatterometer 902, to control the parameter of lithographic equipment 901 effectively.These values of drawing can also be used in the feed-forward signal, thereby can optimally set next process operation.These values of drawing can for example be transported to Etaching device, and Etaching device can be revised the substrate of its setting to be adapted to arrive.
The example of the correctable effect of focal length change gradient in the exposure field, change the skew on the entire wafer and the skew of wafer to wafer.The example that dosage can obtain calibration result change the skew on gradient in the exposure field and/or curvature, the change entire wafer and the skew of wafer to wafer.
According to an embodiment of this method, do not need complicated calculating, do not need to understand the character of substrate yet, just can directly use spectrum to determine the value of at least one technological parameter.And, from spectrum, extracting in the process of relevant information, the regression technique that mathematical model adopted can reduce the influence of noise to desired data.Light detection device can be a scatterometer.Scatterometer configuration is used for rapid and reliable mode measure spectrum, and can be used in a lot of positions on all types of, the substrate that will expose.
According to one embodiment of the invention, can finish measurement on the custom-designed target or on the device pattern in chip area.In another embodiment of the invention, this at least one technological parameter is selected from the group that is made of focal length, exposure dose and aliasing error.Also have the parameter relevant with dosage, as: 1) the guide rail parameter (for example, PEB time/temperature) relevant, the processing operation that promptly has the effect that is similar to the scanner exposure dose with dosage; 2) variation of live width on the whole marking-off plate, or the variation of marking-off plate and marking-off plate.These effects can be proofreaied and correct with exposure dose, and ground can be by model explanation for being exposure dose.Other included technological parameters of this group throw the aberration of camera lens, the angle distribution of glittering and shining the light of mask of projection camera lens, for example ellipticity in addition.In one embodiment of the invention, can be implemented as a parameter or a plurality of parameter difference determined value in these parameters, these parameters are extremely important for dimensional uniformity very crucial in the control photoetching process.
In one embodiment of the invention, the regression technique that mathematical model adopted is selected from down the group that surface technology constitutes: principal component recurrence, the recurrence of non-linear principal component, local LEAST SQUARES MODELS FITTINGization and non-linear local LEAST SQUARES MODELS FITTINGization.
In one embodiment of the invention, operable substrate comprises testing wafer or product wafer.According to concrete application, mark structure can be arranged on any position on the substrate.Thereby mark structure can be positioned in the chip area or be positioned on the groove.In the time of in mark structure is arranged on chip area, mark structure can become the part of device pattern in this chip area.The part that can freely locate diffraction structure or use device architecture has improved the versatility of the inventive method.
In one embodiment of the invention, mark structure comprises diffraction grating.Diffraction grating is a kind of structure that scatterometry is used that is well suited for.
In another embodiment of the invention, this method also is included in uses before the mathematical model calibration data that pre-service obtains and the measurement data that obtains.Pretreated use can improve the performance of mathematical model usually.Be used for pretreated mathematical operations can comprise deduct mean value, divided by standard deviation, selective light mathematic(al) parameter and weighting optical parametric.The example of optical parametric has wavelength, angle and the polarization state of the used light beam of light detection device.
In one embodiment of the invention, at least one in substrate and the calibration substrate comprises at least two different mark structures.In the situation of product substrate, these at least two mark structures can be the product labelling structures, and in the situation of calibration substrate, these at least two mark structures can be the collimating marks structures.For word is oversimplified as far as possible, term used herein " mark structure " refers to both of these case.When adopting pre-service, it is extremely useful using a more than mark structure.These at least two mark structures can be provided with coming in close proximity to each other, so that the distance between these at least two mark structures is identical with the magnitude of mark structure size.
In one embodiment of the invention, these at least two mark structures comprise first mark structure and second mark structure, and wherein, first mark structure comprises several not patterned layers; Second mark structure comprises identical not patterned layer, but is provided with pattern on this layer.In this embodiment, first mark structure is only to the variation sensitivity of patterned layer not.Any spectrum change that causes owing to the variation in the patterned layer not can be detected, and can use when the spectrum that second mark structure is obtained is analyzed.
In another embodiment of the invention, these at least two mark structures comprise first mark structure and second mark structure.Wherein, this first mark structure comprises the pattern with shielding wire, and second mark structure comprises the pattern with intensive line or insulating space.These mark structures can have different susceptibilitys to the technological parameter such as focal length and dosage.As a result, can obtain additional information, these information are useful when determining the value of technological parameter with mathematical model.
In one embodiment, lithographic equipment is connected to guide rail, and light detection device is scatterometer, and it is connected to identical guide rail.Thereby this can provide a kind of effective and efficient manner to monitor and adjust these parameters, makes it to be suitable for the even performance of lithographic equipment.
At least some embodiments of the present invention combine carries out the useful influence that pre-service brings and adopts a more than mark structure to obtain the advantage of information needed the performance of mathematical model.Suitable mathematical operations comprise deduct mean value, divided by standard deviation, selective light mathematic(al) parameter and weighting optical parametric.The example of optical parametric has the used light wavelength of light detection device, angle, polarization state.
In one embodiment of the invention, these at least two collimating marks structures comprise the first collimating marks structure and the second collimating marks structure.Wherein, the first collimating marks structure comprises several not patterned layers; The second collimating marks structure comprises identical not patterned layer, but is provided with pattern on this layer.In this embodiment, the first collimating marks structure is only to the variation sensitivity of patterned layer not.Any spectrum change that causes owing to the variation in the patterned layer not can be detected, and can use when the spectrum that the first collimating marks structure is obtained is analyzed.
In one embodiment of the invention, these at least two collimating marks structures comprise the first collimating marks structure and the second collimating marks structure.Wherein, this first collimating marks structure comprises the pattern with shielding wire, and the second collimating marks structure comprises the pattern with intensive line or insulating space.These mark structures can have different susceptibilitys to the technological parameter as focal length and dosage.As a result, can obtain additional information, these information are useful when determining the value of technological parameter with mathematical model.
In the application of each embodiment of the present invention, off line is implemented calibration usually.In one embodiment of the invention, owing to do not expect to occur the obstruction of production line, therefore just online enforcement is measured after finishing calibration.In one embodiment of the invention, scatterometer can be integrated in the guide rail, so that can carry out on line operation.Alternatively, can measure several substrates of finishing technology, continue simultaneously to handle with scatterometer independently.Yet in the latter's situation, feedback interval increases significantly.When known (for example, by measure) in advance calibrated so and measured and can onlinely carry out when the substrate that is used to calibrate (for example, the product wafer) is gone up the natural variation that exists.
Test findings: in a test, be the two class wafer exposure of 300mm to diameter.
The first kind is smooth calibration chip, and FEM exposes thereon.FEM is by 13 focal length steppings (step-length 30nm) and 9 dosage stepping (step-length 0.5mJ/cm 2, nominal standard dose is approximately 29mJ/cm 2) form.For each structure, use the scatterometer measure spectrum with unique focal length and dose value printing.These spectrum and used focal length and dosage skew combine and generate a regression model.
The second class wafer is a sample wafer, i.e. the wafer that will measure.In this test, two sample wafer have been measured.Two sample wafer all comprise indentation, and indentation deliberately generates in wafer, to obtain more significant focusing effect.A kind of field of setting exposure covering entire wafer with focal length and dosage.Because the natural variation that exists in focal length and the dosage, as mentioned above, the pattern of each exposure will corresponding to slightly different focal length and the dose value of setting value.Subsequently, measure the structure of printing with scatterometer.The regression model that employing is obtained by FEM will belong to each spectral translation coking distance and dose value of the print structure of sample wafer.
The above-mentioned regression model of mentioning is applied to from the spectrum that the sample wafer with indentation obtains, obtains focal length and dose distribution.In order to verify scatterometry result's accuracy, establish mutual relationship with the another kind of method of well setting up.In this test, this correlationship is just set up for dosage, by the result that will obtain with from so-called homogenizing validation test (leveling verification test, LVT) result who obtains compares and sets up this relation, and V.15375-132 this method is for example discussed in (2004) to some extent at people's such as Valley SPIE USE.The line template substrate with wedge shape thickness area is adopted in this test, and this wedge shape thickness area for example forms by a large amount of little prisms is set, and each prism is fixed on and is suitable for measuring on the overlapping mark structure.This line template substrate comprises that also several can be with " common " mark structure for referencial use.The ratio that defocuses of laterally offset and the mark structure below prismatic produces near linear relation at imaging displacement and between defocusing.Thereby focus error changes into aliasing error.After measuring wafer with scatterometer, wafer is stripped from, is also applied and exposure again again, measures to be used for LVT.
The LVT data are inserted in the scatterometry grid.As can be seen from Table 1, exist extraordinary mutual relationship between two kinds of technology.Table 1 has provided with LVT and scatterometry the scatterometry spectrum of two sample wafer and two classes record (has been called α and β 2) variation between the focal length that measures.This mutual relationship is rendered as two kinds of 3 σ-focus difference (dF), regression slope (slope) and related coefficient (R between technology 2).For two wafers, relevant result is very similar, and does not depend on used spectrum types strongly.The upper limit of scatterometry precision is provided by the focus difference between two kinds of technology.Actual precision will be better, and this is because LVT also has certain inaccuracy, and wafer is applied and exposure again again between twice measurement.
DF[nm,3σ] Slope R 2
Wafer 1, α 36 0.88 0.85
Wafer 1, β 36 0.87 0.76
Wafer 2, α 37 0.81 0.81
Wafer 2, β 36 0.81 0.74
Mean value 36 0.85 0.79
Focus difference between table 1 scatterometry and the LVT
In the superincumbent description, what suppose is exposure label(l)ing structure after developing.Yet, also can use potential mark structure, promptly be exposed but also undeveloped mark structure.Potential mark can be soon with regard to imaging after exposure, and this is favourable, because backfeed loop can be faster.In addition, also do not finish because guide rail is handled, so measurement data can be used for the feed-forward signal to guide rail.
Although specific embodiments of the present invention described above is appreciated that the present invention can be to implement with above-mentioned different mode.Embodiment of the present invention (for example also comprise computer program, instruction or instruction sequence are organized in one group of territory more) implement above-mentioned method with the control lithographic equipment, also comprise the storage medium (for example, disk, semiconductor memory) of storing one or more this programs with machine-readable form.Instructions is not to be intended to limit the present invention, and for example, the present invention can be applied in different technical fields, comprises picture photoetching, MRI and radar application and other field.Yet it is particularly advantageous that the present invention is used in the field of lithography, because the complicacy of photoetching and hi-tech make it be difficult to all parameters are controlled to desirable level.The application of the invention can not directly actuated parameter can be controlled indirectly.

Claims (42)

1. method that is used for determining at least one technological parameter, this method comprises:
Obtain the calibration measurement data a plurality of collimating marks structural group on being arranged on calibration target, in described a plurality of collimating marks structural group each all comprises at least one collimating marks structure, and the collimating marks structure in the different collimating marks structural group adopts the different given values of described at least one technological parameter to generate;
Described given value by using described at least one technological parameter is also used regression technique to described calibration measurement data, determines mathematical model, and described mathematical model comprises several regression coefficients;
Obtain measurement data from least one mark structure that is arranged on the target, described at least one mark structure uses the unknown-value of described at least one technological parameter to make; And
By utilizing the described regression coefficient of described mathematical model, from the described measurement data that obtains, determine the unknown-value of described at least one technological parameter for described target.
2. according to the process of claim 1 wherein that described calibration measurement data and described measurement data obtain with photo-detector.
3. according to the method for claim 2, wherein said photo-detector is a scatterometer.
4. according to the method for aforementioned arbitrary claim, wherein the regression technique of mathematical model use is selected from down group, and this group comprises: principal component recurrence, the recurrence of non-linear principal component, local LEAST SQUARES MODELS FITTINGization and non-linear local LEAST SQUARES MODELS FITTINGization.
5. according to the method for aforementioned arbitrary claim, wherein said target is a substrate.
6. according to the method for claim 5, wherein said substrate comprises in the group that is made of testing wafer and product wafer.
7. according to the method for claim 5 or 6, wherein said at least one mark structure is positioned on the described substrate, is arranged in of group that chip area and groove constitute.
8. according to the method for claim 7, wherein said at least one mark structure is a part that is positioned at the device pattern of chip area.
9. according to the method for aforementioned arbitrary claim, wherein at least one mark structure comprises diffraction grating.
10. according to the method for aforementioned arbitrary claim, wherein said method is carried out pre-service to calibration measurement data that obtain and the measurement data that obtains before also being included in and utilizing described regression coefficient.
11. according to the method for claim 10, wherein said pre-service comprises the mathematical operations of described data being implemented in the group down at least one, this group comprises: deduct mean value, divided by standard deviation, selective light mathematic(al) parameter and weighting optical parametric;
Wherein said optical parametric comprises at least one in the one group of parameter that is made of wavelength, angle and polarization state.
12. according to the method for aforementioned arbitrary claim, each of wherein said a plurality of collimating marks structural group all comprises at least the first collimating marks structure and the second different collimating marks structures.
13. according to the method for claim 12, the wherein said first collimating marks structure comprises several not patterned layers, the described second collimating marks structure comprises identical not patterned layer, but is provided with pattern on it.
14. according to the method for claim 12 or 13, the wherein said first collimating marks structure comprises the pattern with shielding wire, the described second collimating marks structure comprises the pattern with intensive line or insulating space.
15. according to any method among the claim 12-14, the wherein said first and second collimating marks structures are closely close each other, so that the distance between the described first and second collimating marks structures has identical magnitude with the size of the described first and second collimating marks structures.
16. according to the method for aforementioned arbitrary claim, wherein at least one calibration structure in the collimating marks structural group and described mark structure have comparable substantially shape.
17. according to the method for aforementioned arbitrary claim, wherein said calibration data and measurement data comprise spectroscopic data.
18. according to the method for aforementioned arbitrary claim, wherein said method is relevant with in lithographic equipment and the guide rail at least one.
19. method according to claim 18, wherein said at least one technological parameter is selected from down group, and this group comprises: the aberration of the variation of variation, marking-off plate and the marking-off plate of live width, projection camera lens, projection camera lens glitters and the angle distribution of the light of the marking-off plate that throw light on focal length, exposure dose, aliasing error, the guide rail parameter relevant with dosage, the marking-off plate.
20. according to the method for claim 18 or 19, wherein said lithographic equipment comprises:
Illuminator, configuration is used to provide radiation beam;
Supporting construction, configuration is used for supporting picture construction, and this picture construction is used for giving pattern to radiation beam in xsect;
The substrate platform, configuration is used for laying substrate; And
Projection system, configuration is used for the light beam of patterning is projected in the target portion of substrate.
21. use the semiconductor devices of making according to the method for aforementioned arbitrary claim.
22. a system that is used for determining at least one technological parameter, this system comprises:
Detector, layout is used for obtaining the calibration measurement data a plurality of collimating marks structural group on being arranged on calibration target, in described a plurality of collimating marks structural group each all comprises at least one collimating marks structure, and the collimating marks structure in the different collimating marks structural group adopts the different given values of described at least one technological parameter to generate;
Processor unit is stored a mathematical model, and this mathematical model determines that by the described given value of described at least one technological parameter of use and to described calibration measurement data application regression technique described mathematical model comprises several regression coefficients;
Described processor unit is arranged and is used for obtaining measured value from least one mark structure that is arranged on the target, described at least one mark structure uses the unknown-value of described at least one technological parameter to make, and be used for from the described measurement data that obtains, determining the unknown-value of described at least one technological parameter for described target by utilizing the described regression coefficient of described mathematical model.
23. according to the system of claim 22, wherein said detector is a photo-detector.
24. according to the system of claim 23, wherein said photo-detector is a scatterometer.
25. according to any system among the claim 22-24, wherein the regression technique of mathematical model use is selected from down group, and this group comprises: principal component recurrence, the recurrence of non-linear principal component, local LEAST SQUARES MODELS FITTINGization and non-linear local LEAST SQUARES MODELS FITTINGization.
26. according to any system among the claim 22-25, wherein said target is a substrate.
27. according to the system of claim 26, wherein said substrate comprises in the group that is made of testing wafer and product wafer.
28. according to the system of claim 26 or 27, wherein said at least one mark structure is positioned on the described substrate, is arranged in of the group that is made of chip area and groove.
29. according to the system of claim 28, wherein said at least one mark structure is a part that is positioned at the device pattern of chip area.
30. according to any system among the claim 22-29, wherein said at least one mark structure comprises diffraction grating.
31. according to any system among the claim 22-30, wherein said processor unit is arranged and is used for before utilizing described regression coefficient the measurement data that obtains being carried out pre-service.
32. according to the system of claim 31, wherein said pre-service comprises the mathematical operations of described data being implemented in the group down at least one, this group comprises: deduct mean value, divided by standard deviation, selective light mathematic(al) parameter and weighting optical parametric;
Wherein said optical parametric comprises at least one that organize down in the parameter, and this group comprises wavelength, angle and polarization state.
33. according to any system among the claim 22-32, each of wherein said a plurality of collimating marks structural group all comprises at least the first collimating marks structure and the second different collimating marks structures.
34. according to the system of claim 33, the wherein said first collimating marks structure comprises several not patterned layers, the described second collimating marks structure comprises identical not patterned layer, but is provided with pattern on it.
35. according to the system of claim 33 or 34, the wherein said first collimating marks structure comprises the pattern with shielding wire, described second mark structure comprises the pattern with intensive line or insulating space.
36. according to any system among the claim 33-35, the wherein said first and second collimating marks structures are closely close each other, so that the distance between the described first and second collimating marks structures has identical magnitude with the size of the described first and second collimating marks structures.
37. according to any system among the claim 22-36, wherein at least one calibration structure in the collimating marks structural group and described mark structure have comparable substantially shape.
38. according to any system among the claim 22-37, wherein said calibration data and measurement data comprise spectroscopic data.
39. according to any system among the claim 22-38, wherein said system comprises at least one in lithographic equipment and the guide rail.
40. system according to claim 39, wherein said at least one technological parameter is selected from down group, and this group comprises: the aberration of the variation of variation, marking-off plate and the marking-off plate of live width, projection camera lens, projection camera lens glitters and the angle distribution of the light of the marking-off plate that throw light on focal length, exposure dose, aliasing error, the guide rail parameter relevant with dosage, the marking-off plate.
41. the system according to claim 39 or 40 comprises:
Illuminator, configuration is used to provide radiation beam;
Supporting construction, configuration is used for supporting picture construction, and this picture construction is used for giving pattern to radiation beam in xsect;
The substrate platform, configuration is used for laying substrate; And
Projection system, configuration is used for the light beam of patterning is projected in the target portion of substrate.
42. use the semiconductor devices of making according to the system of any among the claim 22-41.
CN 200580012771 2004-02-23 2005-02-22 Method to determine the value of process parameters based on scatterometry data Pending CN1947062A (en)

Applications Claiming Priority (3)

Application Number Priority Date Filing Date Title
US54616504P 2004-02-23 2004-02-23
US60/546,165 2004-02-23
US10/853,724 2004-05-26

Publications (1)

Publication Number Publication Date
CN1947062A true CN1947062A (en) 2007-04-11

Family

ID=38045530

Family Applications (1)

Application Number Title Priority Date Filing Date
CN 200580012771 Pending CN1947062A (en) 2004-02-23 2005-02-22 Method to determine the value of process parameters based on scatterometry data

Country Status (1)

Country Link
CN (1) CN1947062A (en)

Cited By (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101510050B (en) * 2009-03-25 2011-09-07 中国科学院微电子研究所 Method for extracting electron beam exposure scattering parameters
CN104583872A (en) * 2012-07-10 2015-04-29 科磊股份有限公司 Model building and analysis engine for combined x-ray and optical metrology
CN105444708A (en) * 2014-09-23 2016-03-30 约翰内斯﹒海德汉博士有限公司 Method for error correction in position measuring devices
CN105765463A (en) * 2013-11-26 2016-07-13 Asml荷兰有限公司 Methods and apparatus for calculating electromagnetic scattering properties of a structure and for estimation of geometrical and material parameters thereof
CN106707697A (en) * 2017-01-04 2017-05-24 上海华虹宏力半导体制造有限公司 Pattern dimension detecting method for chip
CN107039304A (en) * 2016-02-01 2017-08-11 果尼阿克有限公司 Method and apparatus for assessing semiconductor fabrication process
CN108728812A (en) * 2017-04-24 2018-11-02 神华集团有限责任公司 A method of preparing film
CN109690411A (en) * 2016-09-12 2019-04-26 Asml荷兰有限公司 The difference target design and method measured for technique
CN111316169A (en) * 2017-11-01 2020-06-19 Asml荷兰有限公司 Data estimation in metrology
CN111512235A (en) * 2017-12-19 2020-08-07 Asml荷兰有限公司 Calibration and control based on computational measurements
CN112364592A (en) * 2020-11-09 2021-02-12 天津大学合肥创新发展研究院 Silicon-based PIN photoelectric detector modeling method capable of representing technological parameter deviation
US11016396B2 (en) 2017-05-04 2021-05-25 Asml Holding N.V Method, substrate and apparatus to measure performance of optical metrology

Cited By (22)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101510050B (en) * 2009-03-25 2011-09-07 中国科学院微电子研究所 Method for extracting electron beam exposure scattering parameters
CN104583872B (en) * 2012-07-10 2017-04-26 科磊股份有限公司 Model building and analysis engine for combined x-ray and optical metrology
CN104583872A (en) * 2012-07-10 2015-04-29 科磊股份有限公司 Model building and analysis engine for combined x-ray and optical metrology
US10013518B2 (en) 2012-07-10 2018-07-03 Kla-Tencor Corporation Model building and analysis engine for combined X-ray and optical metrology
CN105765463B (en) * 2013-11-26 2018-06-22 Asml荷兰有限公司 Method and apparatus for the electromagnetic scattering property for calculating structure and for estimating its geometry and material parameter
CN105765463A (en) * 2013-11-26 2016-07-13 Asml荷兰有限公司 Methods and apparatus for calculating electromagnetic scattering properties of a structure and for estimation of geometrical and material parameters thereof
US9939250B2 (en) 2013-11-26 2018-04-10 Asml Netherlands B.V. Methods and apparatus for calculating electromagnetic scattering properties of a structure and for estimation of geometrical and material parameters thereof
CN105444708B (en) * 2014-09-23 2018-02-13 约翰内斯﹒海德汉博士有限公司 Method for correcting error in position-measurement device
CN105444708A (en) * 2014-09-23 2016-03-30 约翰内斯﹒海德汉博士有限公司 Method for error correction in position measuring devices
US10060772B2 (en) 2014-09-23 2018-08-28 Dr. Johannes Heidenhain Gmbh Method for correcting errors in position-measuring devices
CN107039304A (en) * 2016-02-01 2017-08-11 果尼阿克有限公司 Method and apparatus for assessing semiconductor fabrication process
CN109690411B (en) * 2016-09-12 2021-12-03 Asml荷兰有限公司 Differential target design and method for process metrology
US12007697B2 (en) 2016-09-12 2024-06-11 Asml Netherlands B.V. Method for process metrology
CN109690411A (en) * 2016-09-12 2019-04-26 Asml荷兰有限公司 The difference target design and method measured for technique
US11385551B2 (en) 2016-09-12 2022-07-12 Asml Netherlands B.V. Method for process metrology
CN106707697A (en) * 2017-01-04 2017-05-24 上海华虹宏力半导体制造有限公司 Pattern dimension detecting method for chip
CN108728812B (en) * 2017-04-24 2020-07-14 国家能源投资集团有限责任公司 Method for preparing film
CN108728812A (en) * 2017-04-24 2018-11-02 神华集团有限责任公司 A method of preparing film
US11016396B2 (en) 2017-05-04 2021-05-25 Asml Holding N.V Method, substrate and apparatus to measure performance of optical metrology
CN111316169A (en) * 2017-11-01 2020-06-19 Asml荷兰有限公司 Data estimation in metrology
CN111512235A (en) * 2017-12-19 2020-08-07 Asml荷兰有限公司 Calibration and control based on computational measurements
CN112364592A (en) * 2020-11-09 2021-02-12 天津大学合肥创新发展研究院 Silicon-based PIN photoelectric detector modeling method capable of representing technological parameter deviation

Similar Documents

Publication Publication Date Title
CN1947062A (en) Method to determine the value of process parameters based on scatterometry data
US9563131B2 (en) Lithographic apparatus, substrate and device manufacturing method
US11106142B2 (en) Metrology recipe selection
EP1721218B1 (en) Method and apparatus to determine the value of process parameters based on scatterometry data, semiconductor device produced therewith
TWI689796B (en) Method of recipe selection, computer program product, and substrate measurment apparatus
US8773657B2 (en) Method to determine the value of process parameters based on scatterometry data
US9915879B2 (en) Substrate and patterning device for use in metrology, metrology method and device manufacturing method
CN102918464B (en) Measurement of a structure on a substrate
KR101848340B1 (en) Methods and apparatus for measuring a property of a substrate
KR101208462B1 (en) Method and apparatus for controlling a lithographic apparatus
CN105892238B (en) Determine the modified method of focal position, photoetching treatment member and device making method
CN107430350B (en) Metering method and equipment, computer program and lithography system
KR101826651B1 (en) Method of determining critical-dimension-related properties, inspection apparatus and device manufacturing method
CN105874389B (en) For designing the method and apparatus for measuring target
CN105934717B (en) It is operable to measure the equipment operated, lithographic equipment and the method that measurement operation is performed to substrate to be performed to substrate
US20110295555A1 (en) Method and System for Determining a Lithographic Process Parameter
CN109478023A (en) Method and apparatus for measuring the design of target field
CN105452963A (en) Method and inspection apparatus and computer program product for assessing a quality of reconstruction of a value of a parameter of interest of a structure
CN1854899A (en) Lithography measurements using scatterometry
US10036962B2 (en) Inspection apparatus and methods, lithographic system and device manufacturing method
CN106062634A (en) Measuring a process parameter for a manufacturing process involving lithography
CN1892431A (en) Method of characterization, method of characterizing a process operation, and device manufacturing method
KR20120080198A (en) Method, inspection apparatus and substrate for determining an approximate structure of an object on the substrate
KR102293144B1 (en) Automatic selection of measurement target measurement recipes
KR102109508B1 (en) Device manufacturing method and associated lithographic apparatus, inspection apparatus, and lithographic processing cell

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C12 Rejection of a patent application after its publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20070411