NEGATIVE ELECTRODE ACTIVE MATERIAL, METHOD FOR MANUFACTURING SAME, AND LITHIUM SECONDARY BATTERY Technical Field [0001] The present invention relates to a negative electrode comprising a lithium-containing silicon oxide and a method for manufacturing the same, arid further relates to a lithium secondary battery using the negative electvtfde. The present invention also relates to an evaluation method for a negative electrode active material. Background Art [0002] Various batterys have been proposed to date as a lithium secondary battery comprising silicon oxide in its negative electrode. For example, Patent Literature 1 describes, as preferable methods for preparing silicon oxide or silicate containing lithium, a synthetic method including mixing simple substance(s) or compound(s) of lithium, silicon, another metallic element or nonmetallic element, and the like in a prescribed molar ratio and heating in the air or an atmosphere containing oxygen, and a method including subjecting silicon oxide to absorb lithium ion thereinto by an electrochemical reaction between silicon oxide, such as silicon dioxide Si02 or silicon monoxide SiO, and lithium or a substance containing lithium (Patent Literature 1, paragraphs 0011 and 0016). [0003] Patent Literatures 2 to 7 describe that the silicon oxide preferably used as an electrode material satisfies that when a spectrum of the silicon oxide is measured by solid state NMR (29SiDD/MAS) with a sufficient relaxation time set, two peaks of a broad peak (Al) centered at -70 ppm and a broad peak (A2) centered at -110 ppm are separately measured, and an area ratio between these peaks (A1/A2) has a value falling in a range of 0.1 < A1/A2 <1.0. Citation List Patent Literature [0004] - 2 - Patent Literature 1: Japanese Patent No. 2997741 Patent Literature 2- Japanese Patent Publication No. 3952118 Patent Literature 3: Japanese Patent Laid-Open No. 2001-216961 Patent Literature 4'- Japanese Patent Publication No. 4752992 Patent Literature 5^ Japanese Patent Publication No. 4288455 Patent Literature 6- Japanese Patent Laid-Open No. 2004-063433 Patent Literature 7- Japanese Patent Laid-Open No. 2009-259723 Summary of Invention Technical Problem [0005] Patent Literatures 2 to 7 describe that silicon oxide is measured by the solid state NMR (29SiDD/MAS) to determine whether it is suitable as a negative electrode active material. However, the measurement was made for the active material in the state containing no lithium, and not for that in a lithium-doped state. [0006] When silicon oxide is used in a negative electrode of a battery, it is doped with lithium through charging. As described in Patent Literature 1 mentioned above, there are various methods for doping silicon oxide with lithium. Whereas, there is a certain limitation to find out a negative electrode active material having high performance by using the evaluation of silicon oxide not containing lithium as described in Patent Literatures 2 to 7. [0007] As a result of investigation made by the present inventors, it was found that the 29Si-DDMAS spectrum of silicon oxide changes when the silicon oxide is doped with lithium and that the change is largely varied depending on a doping amount of lithium and a doping method. Specifically, in the 29Si-DDMAS spectrum of silicon oxide doped with lithium, peaks appear in addition to those in the vicinity of -70 ppm, -84 ppm and -110 ppm described in Patent Literatures 2 to 7, and it is difficult to obtain a battery having high performance unless these peaks are controlled to fall in a proper range. [0008] - 3 - Specifically, an object of the present invention is to provide a negative electrode for a lithium secondary battery and a lithium secondary battery having excellent properties. Solution to Problem [0009] The present is characterized by that a negative electrode active material comprises silicon oxide which satisfies the following Equation 1 and Equation 2 when solid state NMR (29Si-DDMAS) of silicon is measured for the silicon oxide after performing charging at least once, (Equation l) 0.42 < S1/G31 + S2 + S3) < 0.55 (Equation 2) 0.21 < S3/(S1 + S2 + S3) < 0.26, in which Si is a sum of peak areas of a group of signals assigned to Si having a Si'Si bond and having peaks at 0 to -15 ppm, -55 ppm, -84 ppm and -88 ppm, S2 is a sum of peak areas of a group of signals assigned to Si having a Si(OH)4- n(OSi)n (n = 3, 4) structure and having peaks at -100 ppm and -120 ppm, and S3 is a sum of peak areas of a group of signals assigned to Si having a Si(OLi)4- n(OSi)n (n = 0, 1, 2, 3) structure and having peaks at -66 ppm, -74 ppm, -85 ppm and -96 ppm. Advantageous Effects of Invention [0010] According to the present invention, a negative electrode for a lithium secondary battery and a lithium secondary battery having excellent properties can be provided. In addition, according to one aspect of the present invention, a method for manufacturing a negative electrode for a lithium secondary battery having an excellent characteristic can be provided. Brief Description of Drawings [0011] [Figure l] Figure 1 is a schematic cross-sectional view illustrating the structure of a stacked laminate type secondary battery. [Figure 2] Figure 2 is a diagram illustrating a solid NMR (29Si-DDMAS) spectrum of silicon oxide doped with Li and an example of the fitting of the spectrum. - 4 - Description of Embodiment [0012] A method for obtaining sums Si to S3 will be described. Solid state NMR (29Si-DDMAS (Dipolar Decoupling/Magic Angle Spinning)) peaks corresponding to principal Si present in silicon oxide doped with Li (including doping through charging) are divided into the following three groups: [0013] (SI Peak Group) A group of signals having peaks at chemical shifts of 0 to -15 ppm, -55 ppm, -84 ppm and -88 ppm (hereinafter sometimes designated as the SI peak group for simplification). These peaks are assigned to Si having a Si-Si bond. A sum of the peak areas of the SI peak group is defined as the sum SI. [0014] (S2 Peak Group) A group of signals having peaks at chemical shifts of -100 ppm and -120 ppm (hereinafter sometimes designated as the S2 peak group for simplification). These peaks are assigned to Si having a Si(OH)4-n(OSi)n (n = 3, 4) structure. A sum of the peak areas of the S2 peak group is defined as the sum S2. [0015] (S3 Peak Group) A group of signals having peaks at chemical shifts of -66 ppm, -74 ppm, - 85 ppm and -96 ppm (hereinafter sometimes designated as the S3 peak group for simplification). These peaks are assigned to Si having a Si(OLi)4-n(OSi)n (n = 0, 1, 2, 3) structure. A sum of the peak areas of the S3 peak group is defined as the sum S3. [0016] (Determination of Each Peak Area) In a solid state NMR (29Si-DDMAS) spectrum of silicon oxide doped with Li (including doping through charging), the Si to S3 peak groups overlap one another as in an example illustrated in Figure 2. [0017] Therefore, in order to obtain the intensities of the respective peaks, the respective peaks are expressed by using arbitrary Gaussian functions, and - 5 - these Gaussian functions are superposed so as to fit the original data. After the superposition of the original data and the Gaussian functions reach in agreement, the area of each peak can be calculated to obtain the intensity of each peak. It is noted that a spinning side band generated by the MAS (Magic Angle Spinning) is removed from the data before performing the fitting. A spinning side band and a true peak can be distinguished from each other by measuring whether a peak shift is observed when the rotation speed of the MAS is changed. If a peak shift is observed, it is a spinning side band, and if not, it is a true peak. [0018] Among the peaks of the signals defined in the present invention, the positions of the peaks of the signal having a peak at -84 ppm belonging to the SI peak group and -85 ppm belonging to the S3 peak group are extremely close to each other. In performing the fitting, the signal having a peak at -85 ppm belonging to the S3 peak group is preferentially used out of these two peaks, and if the signals have the same concordance rate of the fitting with the original data, the superposition which gives a higher intensity to this peak is preferentially employed. [0019] As software capable of performing such peak search, data fitting and area calculation, for example, Origin (data analysis software manufactured by OriginLab Corporation, see http://www.lightstone.co.jp/origin/pa.htm) can be used. This software has an NLSF (Nonlinear Least Squares Fitter) in which a least-squares method is applied , and can fit a curve having a plurality of arbitrary peaks by using a Gaussian function. [0020] In addition, the solid state NMR measurement is performed to a negative electrode that is constituted by using a negative electrode active material comprising silicon oxide and has been charged at least once. In other words, in the present invention, the performance of a negative electrode active material can be evaluated by measuring the Si-NMR in a state where it has been charged at least once to absorb Li. [0021] Specifically, after constituting a battery by using a negative electrode active material and performing charging at least once, a negative electrode is -j - 6 - taken out as a measurement sample for performing the NMR measurement, and thus, the evaluation can be made. Incidentally, as described in examples later, it is preferable to actually produce a battery and charge it at least once (for example, once) before measuring the NMR, but a simulated battery may be produced instead. [0022] In general, as the performance is degraded through repeated charging/discharging, the NMR of an active material tends to be out of the range of Equation 1 or Equation 2. Accordingly, if t he NMR of silicon oxide that is a negative electrode active material in a charged state after performing the charging/discharging twice or more satisfies Equation 1 and Equation 2, it is estimated that the active material satisfies Equation 1 and Equation 2 also after performing the charging once. Accordingly, in general, a negative electrode having been charged once can be used for measuring the NMR. [0023] Accordingly, the present invention embraces, as the scope of rights, both a negative electrode active material before charging (an active material before charging that will satisfy Equation 1 and Equation 2 when charged at least once) and a negative electrode active material having been charged at least once. In the description given above and below, a "silicon oxide" and a "negative electrode active material" may mean those both before and after charging in some cases, or those either before or after charging in the other cases, which will be apparent from the context. [0024] (Collector) - 12- A metal contained in a collector is preferably a metal that does not form an alloy with Li. Examples of the collector include copper, nickel and alloys thereof. The collector can be in the shape of a foil, a plate or a mesh. [0047] As the collector, a foil or a mesh containing copper as a principal component is particularly preferred. For example, a ratio of copper in the collector is preferably 97 to 100% by mass from the viewpoint of conductivity and heat resistance. [0048] (Negative Electrode Active material) According to the present invention, the negative electrode active material contains at least the silicon oxide satisfying Equation 1 and Equation 2 as described above. Accordingly, the negative electrode active material may contain merely the silicon oxide satisfying Equation 1 and Equation 2, and in addition, may contain a known negative electrode active material in combination. [0049] Specific examples of the active material, to be used in addition to the silicon oxide, include carbon materials such as graphite, coke and hard carbon, lithium alloys such as a lithium-aluminum alloy, a lithium-lead alloy and a lithium-tin alloy, metallic lithium, and a metal oxide having a lower potential than a lithium manganese composite oxide, such as Sn02, SnO, Ti02 or Nl^Os. [0050] If an active material to be used in addition to the silicon oxide is contained, for example, graphite is preferably contained as the negative electrode active material in addition to the silicon oxide satisfying Equation 1 and Equation 2. It is also preferable that the silicon oxide is coated with graphite. [0051] The content of the active material in an active material layer (a layer of the negative electrode excluding the collector) is preferably 40% by mass or more and 99% by mass or less, more preferably 50% by mass or more and 95% by mass or less, and further more preferably 65% by mass or more and 90% by mass or less from the viewpoint of improvement of energy density. [0052] - 1 3 - (Conductivity Imparting Agent) The active material layer may further contain a conductivity imparting agent from the viewpoint of improvement of conductivity. The conductivity imparting agent is not especially limited, and for example, any known conductivity imparting agent may be used. An example of the conductivity imparting agent includes a carbon material. Examples of t he carbon material include graphite, amorphous carbon, diamond-like carbon, carbon black, ketjen black, acetylene black, vapor grown carbon fiber, fullerene, carbon nanotube and composites of these. These conductivity imparting agents may be used alone, or two or more of these may be used together. Incidentally, highly crystalline graphite has high electric conductivity and is good in adhesion to a collector made of a metal such as copper and in voltage flatness. On the other hand, low crystalline amorphous carbon shows comparatively small volume expansion, and hence shows a high effect to reduce the volume expansion of the negative electrode as a whole, and is difficult to be degraded due to ununiformity such as a grain boundary or a defect. [0053] The content of the conductivity imparting agent in the active material layer is preferably 1% by mass or more and 25% by mass or less, more preferably 2% by mass or more and 20% by mass or less, and further more preferably 5% by mass or more and 15% by mass or less. If t he content is 1% by mass or more, sufficient conductivity can be retained. Besides, if the content is 25% by mass or less, the ratio in mass of the active material can be large, and hence, a capacity per mass can be large. [0054] (Binder) A binder is not especially limited, and for example, polyvinylidene fluoride, a vinylidene fluoride-hexafluoropropylene copolymer, a vinylidene fluoride-tetrafluoroethylene copolymer, a styrene-butadiene copolymer rubber, polytetrafluoroethylene, polypropylene, polyethylene, polyimide, polyamide imide and the like can be used. The amount of the binder for the negative electrode to be used is preferably 7 to 20 parts by mass based on 100 parts by mass of the negative electrode active material from the viewpoint of "sufficient binding force" and "high energy" that are in a trade-off relationship. [0055] - 1 4 - From the viewpoint of a binding property with a conductive intermediate layer, the binder is preferably polyimide or polyamide imide. Besides, as a precursor of the binder, a polyamic acid is preferably used, and the same polyamic acid as that used in the conductive intermediate layer is more preferably used. [0056] Now, the constitution of the battery except for the negative electrode will be described. [0057] C o n s t i t u t i n g Elements of Positive Electrode> In the present embodiment, a positive electrode active material is not especially limited as long as it can absorb and desorb a lithium ion respectively in charging and in discharging, and for example, a known active material can be used. The positive electrode active material is preferably a lithium transition metal oxide. The lithium transition metal oxide is not especially limited, and examples include a lithium manganate having a layered structure or a lithium manganate having a spinel structure, such as LiMn02 or LixMn2C>4 (0 < x < 2); LiCoC>2, LiNi02, or any of these in which a part of the transition metal is replaced with another metal; a lithium transition metal oxide in which a specific transition metal does not exceed half the number of transition metals, such as LiNii/3Coi/3Mni/302! one having an olivine structure, such as LiFePO^ and these lithium transition metal oxides in which Li is more excessive than in a stoichiometric composition. In particular, LiaNipCoyAl502 (l < a < 1.2, p + y + 8 = 1, p > 0.7, y < 0.2) or LiaNipCoyMns02 (l < a < 1.2, p + y + 8 = l, p> 0.6, Y ^ 0.2) is preferred. These materials can be used alone, or in combinations of two or more. [0058] The positive electrode of t he present embodiment may also contain, in addition to the positive electrode active material, a positive electrode conductivity imparting agent and a positive electrode binder. [0059] As the positive electrode conductivity imparting agent, the carbon materials mentioned above as the examples of the negative electrode conductivity imparting agent, a metallic substance such as aluminum, a powder of a conductive oxide and the like can be used. - 15- [0060] The positive electrode binder is not especially limited, and for example, polyvinylidene fluoride, a vinylidene fluoride-hexafluoropropylene copolymer, a vinylidene fluoride-tetrafluoroethylene copolymer, a styrene-butadiene copolymer rubber, polytetrafluoroethylene, polypropylene, polyethylene, polyimide, polyamide imide and the like can be used. Among these, polyvinylidene fluoride (PVdF) is preferably used from the viewpoint of versatility and low cost. [0061] The content of the positive electrode binder in a positive electrode active material layer is preferably 1% by mass or more and 25% by mass or less, more preferably 2% by mass or more and 20% by mass or less, and further more preferably 5% by mass or more and 15% by mass or less. If the content is 1% by mass or more, the occurrence of electrode peeling can be prevented. Besides, if the content is 25% by mass or less, the ratio in mass of the positive electrode active material can be large, and hence, a capacity per mass can be large. [0062] As a positive electrode collector, from the viewpoint of electrochemical stability, nickel, copper, silver, aluminum and an alloy of these are preferred. Examples of the shape include a foil, a plate and a mesh. In particular, an aluminum foil is preferably used. [0063] To the positive electrode active material layer containing the positive electrode active material, a conduction aid may be added for purpose of lowering impedance. An example of the conduction aid includes a carbonaceous fine particle such as graphite, carbon black or acetylene black. [0064] The positive electrode can be produced, for example, by preparing a positive electrode slurry by mixing a lithium manganese composite oxide, a conductivity imparting agent and a positive electrode binder, and applying the positive electrode slurry on a positive electrode collector. [0065] As an electrolyte, for example, a liquid electrolyte (an electrolyte solution) can be used. In a preferable embodiment, although not especially limited, for - 16- example, an electrolyte solution containing an electrolyte salt and a nonaqueous electrolyte solvent is used. [0066] The nonaqueous electrolyte solvent is not especially limited, and from the viewpoint of stability at a metallic lithium potential, examples include cyclic carbonates such as propylene carbonate, ethylene carbonate, butylene carbonate and vinylene carbonate! chain carbonates such as dimethyl carbonate, diethyl carbonate, ethyl methyl carbonate and dipropyl carbonate! and lactones such as y-butyrolactone. The nonaqueous electrolyte solutions can be used alone, or in combinations of two or more. [0067] The electrolyte salt is not especially limited, and examples include lithium salts such as LiPFG, LiAsF6, LiAlCU, LiC104, LiBF4, LiSbF6, LiCF3S03, LiCF3C02, Li(CF3S02)2 and LiN(CF3S02)2. These electrolyte salts can be used alone, or in combinations of two or more. [0068] Alternatively, an ionic liquid can be used as the electrolyte solution. An example of the ionic liquid includes a quaternary ammonium-imide salt. [0069] Alternatively, a solid electrolyte may be used instead of the liquid electrolyte. Examples of the solid electrolyte include a gel electrolyte obtained by impregnating, with any of the above-described electrolyte solutions, a polymer such as polyacrylonitrile or polyacrylate, and a solid electrolyte such as LiPON or Li2S-LiPxOy (x = 1 to 2, y = 2 to 4). [0070] A separator is not especially limited, and for example, a known separator can be employed. As the separator, for example, a porous film or a nonwoven fabric of polypropylene, polyethylene or the like can be used. Alternatively, a film of polyimide or aramid, a film of cellulose, or the like can be used. [0071] An outer package is not especially limited, and any outer package can be used as long as it is stable against an electrolyte solution and has a sufficient water vapor barrier property. As the outer package, for example, a metal can - 1 7 - of iron, an aluminum alloy or the like, a laminate film, or the like can be used. As the laminate film, a laminate film on which aluminum or silica is vapor deposited is preferably used from the viewpoint of the water vapor barrier property. [0072] Constitution of Battery> The constitution of the secondary battery of the present embodiment is not especially limited, and for example, a constitution in which an electrode element including a positive electrode and a negative electrode opposing each other, and an electrolyte solution are packed in an outer package can be employed. The shape of the secondary battery is not especially limited, and examples include a cylindrical type, a flat rolled rectangular type, a stacked rectangular type, a coin type, a flat rolled laminate type, and a stacked laminate type. [0073] Now, a stacked laminate type secondary battery will be described as an example. Figure 1 is a schematic cross-sectional view illustrating the structure of an electrode element included in a stacked type secondary battery using a laminate film as the outer package. This electrode element is formed by alternately stacking a plurality of positive electrodes c and a plurality of negative electrodes a with a separator b sandwiched therebetween. Positive electrode collectors e of the respective positive electrodes c are mutually welded at their ends not covered by a positive electrode active material to be electrically connected to one another, and a positive terminal f is further welded to the welded portion. Negative electrode collectors d of the respective negative electrodes a are mutually welded at their ends not covered by a negative electrode active material to be electrically connected to one another, and a negative terminal g is further welded to the welded portion. [0074] The electrode element having such a planar stacked structure does not have a portion with a small R (a region close to a rolling core of a coaxially rolled structure, or a folded region corresponding to an end of a flat rolled structure), and therefore, has an advantage that this electrode element is difficult to be harmfully affected in the volume change of the electrode through charging/discharging as compared with an electrode element having a rolled - 1 8 - structure. In other words, it is effectively used as an electrode element using an active material easily causing the volume expansion. On the other hand, since an electrode is bent in an electrode element having a rolled structure, the structure of the electrode element is easily deformed when the volume change occurs. In particular, if a negative electrode active material that undergoes large volume change through the charging/discharging, such as silicon oxide, is used, it is presumed that the capacity is largely lowered through the charging/discharging in a secondary battery using an electrode element having a rolled structure. Examples [0075] Specific examples of the present embodiment will now be described, and it should be noted that the present embodiment is not limited to these examples. [0076] [Example 1] (Preparation of Negative Electrode) [0077] SiO (trade name^ "SIO05PB", manufactured by Kojundo Chemical Laboratory Co., Ltd.), carbon black (trade name: "#3030B", manufactured by Mitsubishi Chemical Corporation) and a polyamic acid (trade name: "U-Varnish A", manufactured by Ube Industries, Ltd.) were weighed at a mass ratio of 80:5:15. These were mixed with N-methylpyrrolidone (NMP) by using a homogenizer to give a slurry. A mass ratio between the NMP and a solid content was set to 57:43. The slurry was applied on a copper foil by using a doctor blade. Thereafter, the resultant was heated at 120°C for 7 minutes for drying the NMP. Then, the resultant was heated at 350°C for 30 minutes under a nitrogen atmosphere by using an electric furnace, and thus, a negative electrode was prepared. [0078] A lithium foil was brought into contact with the negative electrode and kept at 85°C for 8 hours under a nitrogen atmosphere. On the basis of the weight change of the negative electrode, the amount of lithium doped in the negative electrode was obtained. The amount of lithium doped per mol of SiO was 1.62 mol. - 19- [0079] Thereafter, the negative electrode was electrochemically doped with lithium by using a beaker cell. A metallic lithium foil was used as a counter electrode of the beaker cell, and a solution containing 1.0 mol/1 of LiPFe and a mixed solvent (7:3 (in a volume ratio)), as a nonaqueous electrolyte solvent, of ethylene carbonate and diethyl carbonate was used as an electrolyte solution. The amount of lithium doped in the negative electrode was obtained on the basis of a Coulomb amount. The amount of lithium doped per mol of SiO was 1.78 mol. [0080] (Preparation of Positive Electrode) Lithium cobaltate (manufactured by Nichia Corporation), carbon black (trade name: "#3030B", manufactured by Mitsubishi Chemical Corporation) and polyvinylidene fluoride (trade name: "#2400", manufactured by Kureha Corporation) were weighed at a mass ratio of 95:2:3. These were mixed with NMP to give a slurry. The mass ratio between the NMP and a solid content was set to 52:48. The slurry was applied to an aluminum foil with a thickness of 15 um by using a doctor blade. The aluminum foil on which the slurry was applied was heated at 120°C for 5 minutes for drying the NMP, and thus, a positive electrode was prepared. [0081] (Assembly of Secondary Battery) An aluminum terminal and a nickel terminal were welded respectively to the positive electrode and the negative electrode thus prepared. These were stacked with separators sandwiched therebetween to prepare an electrode element. Incidentally, the masses of the positive electrode and the negative electrode were adjusted so that the amount of lithium doped in the SiO, that is, the negative electrode active material, could be an arbitrary value when the battery was in a fully charged state. A nickel terminal was welded to a reference electrode prepared by adhering a copper foil to metallic lithium, and the resultant was stacked on the negative electrode with a separator sandwiched therebetween. The electrode element and the reference electrode were packed in a laminate film, and an electrolyte solution was injected to the inside of the laminate film. Thereafter, the laminate film was sealed by heat bonding while reducing the pressure within the laminate film. In this manner, -20- a planar secondary battery before initial charging was prepared. Herein, a polypropylene film was used as the separator. As the laminate film, a polypropylene film on which aluminum was vapor deposited was used. As the electrolyte solution, a solution containing 1.10 mol/1 of LiPFe as an electrolyte and a mixed solvent (in 7^3 (in a volume ratio)) of ethylene carbonate and diethyl carbonate as a nonaqueous electrolyte solvent was used. [0082] (Charge/discharge Cycle Test of Secondary Battery) The secondary battery prepared as described above was subjected to a charge/discharge cycle test in a range of a battery voltage of 2.5 to 4.2 V. The charging was performed by a CCCV method, and after a voltage of 4.2 V was attained, the voltage was kept constant for 1 hour. The discharging was performed by a CC method (at a constant current of 0.2 C). Here, the 0.2 C current means a current at which it takes 5 hours to completely discharge an arbitrary battery in a fully charged state if the battery is discharged at a constant current. The results of initial charge/discharge efficiency and a cycle characteristic are shown in Table 1. [0083] (NMR Measurement) The secondary battery prepared as described above was charged once, and then, the battery was disassembled in an argon atmosphere to take out the negative electrode, and a negative electrode layer peeled off from the collector was put in a sample tube to be used as an NMR measurement sample. This sample was used for measuring solid state NMR (29SiDD/MAS), so that peaks could be analyzed by the method described above to obtain areas of S i , S2 and S3. The results of Sl/(S1 + S2 + S3) and S3/(S1 + S2 + S3) are shown in Table 1. [0084] [Example 2] A heating time employed for the thermal doping of lithium in a negative electrode was set to 6 hours. The resultant doping amount was 1.39 mol. In addition, an amount of lithium doped attained by using a beaker battery was changed to 2.01 mol. Except for these, a battery was prepared and evaluated in the same manner as in Example 1. [0085] - 2 1 - [Example 3] A heating time employed for the thermal doping of lithium in a negative electrode was set to 4 hours. The resultant doping amount was 1.21 mol. In addition, an amount of lithium doped attained by using a beaker battery was changed to 2.19 mol. Except for these, a battery was prepared and evaluated in the same manner as in Example 1. [0086] [Example 4] A heating time employed for the thermal doping of lithium in a negative electrode was set to 2 hours. The resultant doping amount was 1.02 mol. In addition, an amount of lithium doped attained by using a beaker battery was changed to 2.38 mol. Except for these, a battery was prepared and evaluated in the same manner as in Example 1. [0087] [Comparative Example l] A heating time employed for the thermal doping of lithium in a negative electrode was set to 18 hours. The resultant doping amount was 3.14 mol. The lithium doping by using a beaker battery was not performed. Except for these, a battery was prepared and evaluated in the same manner as in Example 1. [0088] [Comparative Example 2] The thermal lithium doping of a negative electrode was not performed. In addition, an amount of lithium doped attained by using a beaker battery was changed to 3.40 mol. Except for these, a battery was prepared and evaluated in the same manner as in Example 1. [0089] [Comparative Example 3] Neither the thermal lithium doping nor the lithium doping by using a beaker battery in a negative electrode was performed. Except for these, a battery was prepared and evaluated in the same manner as in Example 1. -23- [0091] It is understood from the results shown in Table 1 that the initial charge/discharge efficiency is high but the capacity retention ratio after 200 cycles is low in Comparative Examples 1 and 2, and that the capacity retention ratio after 200 cycles is high but the initial charge/discharge efficiency is low in Comparative Example 3. In all the examples, both the initial charge/discharge efficiency and the capacity retention ratio after 200 cycles are high, and thus, it was revealed that the present invention provides a secondary battery having a high energy density and a good charge/discharge cycle characteristic. In addition, it is understood, based on peaks of the NMR, that the ratio of the silicon having a Si-Si bond is higher in Comparative Examples 1 and 2 than in Comparative Example 3. On the contrary, in all the examples, the ratio of the silicon having a Si-Si bond is suppressed to the same extent as in Comparative Example 3. It seems that since the examples are successful in increasing the ratio of the silicon having the Si(OLi)4-n(OSi)n (n = 0, 1, 2, 3) structure while suppressing the ratio of the silicon having a Si-Si bond to be low, both the charge/discharge efficiency and the capacity retention ratio can be increased. Claims 1. A negative electrode active material comprising silicon oxide which satisfies the following Equation 1 and Equation 2 when solid state NMR (29Si-DDMAS) of silicon is measured for the silicon oxide after performing charging at least once, (Equation l) 0.42 < Sl/(S1 + S2 + S3) < 0.55 (Equation 2) 0.21 < S3/(S1 + S2 + S3) < 0.26, in which Si is a sum of peak areas of a group of signals assigned to Si having a Si-Si bond and having peaks at 0 to -15 ppm, -55 ppm, -84 ppm and -88 ppm, S2 is a sum of peak areas of a group of signals assigned to Si having a Si(0H)4-n(0Si)n (n = 3, 4) structure and having peaks at -100 ppm and -120 ppm, and S3 is a sum of peak areas of a group of signals assigned to Si having a Si(OLi)4-n(OSi)„ (n = 0, 1, 2, 3) structure and having peaks at -66 ppm, -74 ppm, -85 ppm and -96 ppm. 2. The negative electrode active material, wherein conditions of the Equation 1 and Equation 2 are satisfied after performing the charging once. 3. The negative electrode active material according to claim 1 or 2, wherein the negative electrode active material contains silicon oxide predoped with lithium, and conditions of the Equation 1 and Equation 2 are satisfied after performing the charging at least once. 4. The negative electrode active material according to claim 3, wherein the silicon oxide is pre-doped with lithium by employing at least one of (i) a method utilizing thermal diffusion of lithium and (ii) a method of electrochemically doping lithium. 5. The negative electrode active material according to claim 4, wherein after performing pre-doping by (i) utilizing thermal diffusion of lithium, the pre-doping is performed by (ii) the method of electrochemically doping lithium. - 2 5 - 6. A lithium secondary battery comprising a negative electrode comprising a negative electrode active material according to at least one of claims 1 to 5, a binder and a collector, a positive electrode and an electrolyte. 7. A method for manufacturing a negative electrode active material comprising silicon oxide, comprising a step of doping the silicon oxide with lithium, wherein lithium is doped in the step of doping the silicon oxide with lithium so as to satisfies the following Equation 1 and Equation 2 when solid state NMR (29Si-DDMAS) of silicon is measured for the silicon oxide after performing charging at least once, (Equation l) 0.42 < S1/(S1 + S2 + S3) < 0.55 (Equation 2) 0.21 < S3/(S1 + S2 + S3) < 0.26, in which Si is a sum of peak areas of a group of signals assigned to Si having a Si-Si bond and having peaks at 0 to -15 ppm, -55 ppm, -84 ppm and -88 ppm, S2 is a sum of peak areas of a group of signals assigned to Si having a Si(OH)4-n(OSi)n (n = 3, 4) structure and having peaks at -100 ppm and -120 ppm, and S3 is a sum of peak areas of a group of signals assigned to Si having a Si(OLi)4-n(OSi)n (n = 0, 1, % 3) structure and having peaks at -66 ppm, -74 ppm, -85 ppm and -96 ppm. 8. The method for manufacturing a negative electrode active material according to claim 7, wherein the silicon oxide is doped with lithium by employing at least one of (i) a method utilizing thermal diffusion of lithium and (ii) a method of electrochemically doping lithium. 9. The method for manufacturing a negative electrode active material according to claim 8, wherein the silicon oxide is doped with lithium by (i) the method utilizing thermal diffusion of lithium and thereafter doped with lithium by (ii) the method of electrochemically doping lithium. 10. An evaluation method for a negative electrode active material comprising silicon oxide, the method comprising: measuring solid state NMR (29Si-DDMAS) of silicon for the silicon oxide after performing charging at least once, and selecting the negative electrode - 2 6 - active material depending on whether the following Equation 1 and Equation 2 is satisfied, , (Equation l) 0.42 < S1/(S1"+ S2 4- S3) < 0.55 (Equation 2) 0.21 < S3/(S1>+ S2 + S3) < 0.26, in which SI is a sum of peak areas of a group of signals assigned to Si having a Si-Si bond and having peaks at 0 to -15 ppm, -55 ppm, -84 ppm and -88 ppni, S2 is a sum of peak areas of a group of signals assigned to Si having a Si(OH)4-n(OSi)n (n - 3, 4) structure and having peaks at -100 ppm and -120 ppm, and S3 is a sum of peak areas of a group of signals assigned to Si having a Si(OLi)4-n(OSi)n (n = 0, 1, 2, 3) structure and having peaks at -66 ppm, -74 ppm, -85 ppm and -96 ppm.