A kind of method of measuring the deep trench inefficacy degree of depth
Technical field
The present invention relates to the analysis and detection technology in the ic manufacturing process, especially relate to a kind of method of measuring the deep trench inefficacy degree of depth.
Background technology
In recent years, along with the development of semiconductor fabrication process, cross-section structure is increasing to the influence of the yield of high-end device and reliability.For example, deep trench (Deep Trench, DT) in, the shape of sidewall, size etc. are the key factors that influences performance of semiconductor device.
Be to improve the quality of the deep trench course of processing, need analyze, the data of analyzing gained are fed back to production division the inefficacy degree of depth of the groove of the position of the deep trench that lost efficacy and inefficacy.This be because, the course of processing of deep trench need be divided into a plurality of stages usually carries out, the process conditions in each stage (for example gas mixing ratio etc.) all need to regulate accurately and control, in case deviation occurs, then can cause the decreased performance of product.Therefore, the degree of depth of the inefficacy groove that need obtain according to analysis is judged process conditions process segment improperly, thereby perfect in addition.
The position of the deep trench that lost efficacy can be measured by MOSAID by FA (Fail Analysis, failure analysis) assay laboratory, further measures the inefficacy degree of depth of this inefficacy deep trench subsequently.
Seeing also Fig. 1, is a kind of schematic diagram of measuring the method for the deep trench inefficacy degree of depth of the prior art.
At first, sample is placed in FIB (Focus Ion Beam, focused ion beam)/SEM (ScanningElectron Microscope, scanning electron microscopy) system.
The principle of FIB is to utilize high energy ion beam (I-beam) the bombardment sample surfaces that focuses on through electromagnetic lens, the atom sputtering of sample is come out, by long-time sputter, dig out the groove at certain depth and certain inclination angle in the appointed area that needs are observed with ion beam, so that adopt SEM to observe.The prior art that adopts FIB to carry out sample making can be consulted Chinese patent No. 200310122898.1 and No. the 200410075149.2nd, Chinese patent.
Subsequently, the employing ion beam cuts out triangle space 920 near inefficacy groove 910.
After this, adopt ion beam to carrying out cutting progressively near the sample of inefficacy groove 910 from far near, and cooperate the view of the sidewall of electron beam (E-beam) observation sample, until noting abnormalities.
But there are some defectives in the method for the mensuration deep trench inefficacy degree of depth of the prior art.
At first, the method for the prior art need expend the more time.This be because, though cut out the use that triangle space 920 can reduce ion beam,, because still there is certain width in sample between triangle space 920 and inefficacy groove 910, need cut step by step and observe.That is to say, after cutting out the degree of depth of 8 μ m downwards, need to adopt electron beam to observe at every turn, and then the direction moving iron bundle of close inefficacy groove 910, approximately need to move 40 to 50 times, and ion beam whenever the distance of mobile 5nm need about 1 minute consuming time.
Next, the analysis that the method for employing the prior art is carried out is difficult for successfully.This be because, the integral width of deep trench is about about 100nm, and the mobile approximately at every turn 5nm of ion beam, thus the phenomenon that failure site is cut away when often appearing at cutting, thereby can't measure accurately and analyze the inefficacy degree of depth.And the overall depth of deep trench is about about 8 μ m, and ion beam is difficult to control its collimation in downward cutting process, causes the cross section injustice that cuts out, thereby influences the analysis of pair cross-section view.
In view of this, a kind of method of the inefficacy degree of depth of novel mensuration deep trench need be provided.
Summary of the invention
At above-mentioned defective, the problem to be solved in the present invention provides a kind of method of measuring the deep trench inefficacy degree of depth, can save time and improve the accuracy and the success rate of analysis.
In order to address the above problem, the technical solution used in the present invention is: a kind of method of measuring the deep trench inefficacy degree of depth is provided, comprises step:
1) prepares the position of deep trench in sample that sample and acquisition were lost efficacy;
2) when the focused ion beam sample stage turns to predetermined angular, adopt ion beam to cut out pit perpendicular to sample surfaces at a side preset distance place of deep trench;
3) rotate the focused ion beam sample stage, employing comprises the sample of inefficacy deep trench with the ion beam cutting that sample surfaces is inclined to predetermined angular, and adopts and watch the top graph picture of inefficacy deep trench perpendicular to the electron beam of sample surfaces.
Preferably, in described step 3), further look like to regulate the energy of ion beam according to the top graph of inefficacy deep trench.
Preferably, in described step 2) in, further the opposite side in deep trench adopts the ion beam perpendicular to sample surfaces to cut out pit.
Preferably, the degree of depth of described pit is identical with the degree of depth of deep trench, and the ratio of its width and the degree of depth is in 1/2 to 1 the scope.
Preferably, in the described step 1), when preparing sample specimen surface handled can locate failure layer.
Preferably, described surface treatment is to adopt 49% hydrofluoric acid HF to soak sample about 3 minutes.
Preferably, if in deep trench, be filled with polymer, before described step 3), also comprise test piece is handled so that resolution image.
Preferably, described processing to test piece adopts IEE gas to carry out.
Preferably, described step 2) and the predetermined angular of step 3) be 52 the degree.
Preferably, the degree of depth of described deep trench is 8 μ m, and pit is 3 μ m apart from deep trench, and the width of pit is 6 μ m.
Compared with prior art, the invention has the beneficial effects as follows: because the present invention adopts SEM to observe the top graph picture of inefficacy deep trench, therefore, prior art is observed the sidepiece image relatively, and accuracy and success rate are all promoted.And can in cutting, observe, reduce the time of analyzing.In addition, owing to be not from as far as closely sample being carried out perpendicular cuts near the inefficacy deep trench, but directly use predetermined angle tilt cutting to comprise the sample of inefficacy deep trench, can reduce employed clipping time, and be unlikely to cut away failure site, further make accuracy and success rate more reliable.
In preferred version of the present invention, when sample is cut, the energy of ion beam is regulated according to actual conditions, further reduce the time and reduce cost.
In preferred version of the present invention, all cut out pit in the both sides of inefficacy deep trench, make that near the width of the sample that deep trench is is shorter relatively, therefore cut employed ion beam and can reduce, and relative simple to the control of ion beam with adjusting.
Description of drawings
Fig. 1 is a kind of schematic diagram of measuring the method for the deep trench inefficacy degree of depth of the prior art;
Fig. 2 is the flow chart of method of the inefficacy degree of depth of mensuration deep trench of the present invention;
Fig. 3 is the schematic diagram that cuts out pit at the side employing ion beam of the deep trench that lost efficacy;
Fig. 4 is the schematic diagram that cuts out pit at the opposite side employing ion beam of the deep trench that lost efficacy;
Fig. 5 cuts and watches the schematic diagram of top view to the deep trench that lost efficacy;
Fig. 6 is the example that adopts the image of the inefficacy that method of the present invention observes;
Fig. 7 is the schematic diagram that adopts the inefficacy image that another embodiment of method of the present invention observes;
Fig. 8 is the schematic diagram of the another embodiment of method of the present invention.
Embodiment
Seeing also Fig. 2, is the flow chart of method of the inefficacy degree of depth of mensuration deep trench of the present invention.
Step S210, sample is prepared.
In the embodiments of the invention, the size of downcutting sample is about 10mm*10mm, and surface treatment is carried out in the sample test piece, makes and can carry out processing described later to the deep trench layer.
Wherein, this surface-treated process can adopt the method for chemistry, for example adopts 49% hydrofluoric acid HF to soak about 3 minutes, makes the test piece surface treatment to that one deck that can the locate failure address.
Step S220 puts into FIB with sample, rotates the FIB sample stage to predetermined angular, makes ion beam perpendicular to sample surfaces, measures the position of the deep trench that lost efficacy.
In the embodiments of the invention, described predetermined angular is 50 degree.Certainly, at different FIB, the angle that its sample stage is rotated can be different.
The method that those skilled in the art will recognize that the position of measuring the inefficacy groove can be any technological means of prior art, and this does not give unnecessary details.
Step S230 cuts out pit in the precalculated position, both sides of the deep trench that lost efficacy.
See also Fig. 3 and Fig. 4, adopt ion beam successively to cut out pit 120 and 130 in the both sides of the deep trench 110 that lost efficacy.
In one embodiment of the present of invention, the degree of depth of deep trench 110 is about 8 μ m, and pit 120 and 130 is positioned at the both sides of deep trench 110 and apart from deep trench 110 about 3 μ m.Pit 120 is identical with 130 width, is about about 6 μ m.Pit 120 is identical with 130 the degree of depth, and equals the degree of depth of deep trench 110, is about about 8 μ m.
The effect of cutting pit is, reduces the use complexity of ion beam as far as possible, reduces the time of whole analysis.This be because, cut out pit after, shorter relatively near the width of the sample the deep trench, therefore cut employed ion beam and can reduce, and to the control of ion beam with regulate relative simple.And the width of pit is bigger, can adopt thicker ion beam to cut, because the energy of ion beam improves, the time of cutting is minimized, and is about about 15 seconds.
Need to prove that the distance of so-called pit, width and the degree of depth etc. can be adjusted according to actual conditions and demand, are not limited to the described numerical value of present embodiment.Wherein, the distance of pit and deep trench and the width of pit should not have influence on the analysis to deep trench, and are convenient to the operation of ion beam.The width of pit and the ratio of the degree of depth are preferably in 1/2 to 1 the scope.
Step S240, cutting comprises the sample of inefficacy groove, observes the top graph picture of inefficacy groove.
See also Fig. 5, rotate the FIB sample stage, make the Surface Vertical of sample form predetermined angle in electron beam and with ion beam.And different, in one embodiment of the present of invention, this angle is 0 degree to described predetermined angular according to the difference of FIB system.
Adopt the interior sample of ion beam cutting inefficacy groove position range of FIB, adopt the while electron beam to watch Top view (top view) the SEM image of this inefficacy deep trench.
The limit cutting edge is watched, until the inefficacy degree of depth place that is cut to inefficacy deep trench 110.Because in the present embodiment, the Surface Vertical of sample forms predetermined angle in electron beam and with ion beam, can realize that therefore the limit cutting edge watches, thereby reduce the time of analyzing.
Seeing also Fig. 6, is the example that adopts the image of the inefficacy that method of the present invention observes.
Because the present invention adopts SEM to observe the top graph picture of inefficacy deep trench, therefore, prior art is observed the sidepiece image relatively, and accuracy and success rate are all promoted.And can in cutting, observe, reduce the time of analyzing.In addition, owing to be not from as far as closely sample being carried out perpendicular cuts near the inefficacy deep trench, but directly use predetermined angle tilt cutting to comprise the sample of inefficacy deep trench, can reduce employed clipping time, and be unlikely to cut away failure site, further make accuracy and success rate more reliable.
In addition, need to prove,, then, need before watching, handle test piece in order successfully to adopt electron beam to watch the SEM image of sample if in deep trench, be filled with Poly (polymer).
In one embodiment of the present of invention, described processing is to adopt IEE (Insulator Enhance Etch, insulation strengthens etching) gas to carry out.Certainly, described processing also can adopt other modes of prior art to carry out, and this does not give unnecessary details.
Because the comparative between polymer and the sample is not strong, if test piece is not handled, then can't differentiate accurately when watching.And the effect that test piece is handled just is to strengthen contrast between the two, so that differentiate clearly.
In addition, among the present invention, when sample is cut, can also regulate the energy of ion beam, further reduce the time and reduce cost according to actual conditions.
See also Fig. 7, the image of the time watching when cutting is shown in 710 the time, and expression has on failpoint, and then the energy of ion beam can suitably strengthen, the time of accelerating to cut.When the image of watching was shown in 720, then expression had arrived near the inefficacy degree of depth, therefore suitably reduced the energy of ion beam, till the image of watching shown in 730.
See also Fig. 8, in an alternative embodiment of the invention, according to the actual requirements, only the side in inefficacy deep trench position cuts out pit, rather than all cuts out pit in both sides.
In sum, the present invention is when analyzing the inefficacy sample, on the basis of the position of measuring the inefficacy deep trench, further adopt a kind of abrasive method that can fix a point, whole deep trench progressively ground from pushing up downwards, whether the top view (Top view) of observing deep trench while grinding has problem, thereby overcomes the defective of prior art.
The above only is a preferred implementation of the present invention; should be pointed out that for those skilled in the art, under the prerequisite that does not break away from the principle of the invention; can also make some improvements and modifications, these improvements and modifications also should be considered as protection scope of the present invention.