JP2005216487A - Manufacturing method of stamper - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a medium for recording optical information capable of satisfactorily recording a short mark at high speed and having mark stability, and more specifically to provide a manufacturing method of a stamper which can be used for the medium for recording optical information. <P>SOLUTION: The manufacturing method of the stamper to which a phase modulated wobble signal is imparted by groove wobbling has a step for performing exposure to a photoresist layer formed on a glass original plate by wobbling a laser beam for photoresist exposure in a radial direction according to a binary phase modulated wobbling waveform between voltages of ±Vw, wherein the wobbling of the laser beam for photoresist exposure is performed by applying a ring modulator output wave to an EO modulator. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to a method of manufacturing a stamper that can be used for an optical recording medium for high density recording having a phase change recording layer such as a rewritable DVD.

In general, in a compact disc (CD) or DVD, recording of a binary signal and detection of a tracking signal are performed using a reflectance change caused by interference of reflected light from the bottom and mirror surface of a concave pit.
In recent years, phase-change rewritable compact discs (CD-RW, CD-Rewritable) are being widely used as media compatible with CDs. As for DVDs, various phase-changeable rewritable DVDs have been proposed.

These phase-changeable rewritable CDs and DVDs detect a recorded information signal by using a reflectance difference and a phase difference change caused by a difference in refractive index between an amorphous state and a crystalline state. A typical phase change medium has a structure in which a lower protective layer, a phase change recording layer, an upper protective layer, and a reflective layer are provided on a substrate, and a reflectance difference and a phase difference are obtained by using multiple interference of these layers. It can be controlled and compatible with CDs and DVDs.
In CD-RW, the compatibility of CD with recording signals and groove signals can be ensured within the range where the reflectance is reduced to 15 to 25%, and playback is performed with a CD drive to which an amplification system that covers low reflectance is added. Is possible.

Since the phase change recording medium can perform erasing and re-recording processes only by intensity modulation of one focused light beam, recording in a phase change recording medium such as a CD-RW or a rewritable DVD is a recording. And overwrite recording that performs erasure simultaneously.
For recording information using phase change, crystalline, amorphous, or a mixed state thereof can be used, and a plurality of crystal phases can be used. In general, recording media are recorded in an amorphous state by forming an unrecorded / erased state in a crystalline state. As the material for the recording layer, a chalcogen element, that is, a chalcogenide alloy containing S, Se, or Te is often used.

For example, a GeSbTe system mainly composed of GeTe-Sb ₂ Te ₃ pseudo binary alloy, an InSbTe system mainly composed of InTe-Sb ₂ Te ₃ pseudo binary alloy, and an eutectic system composed mainly of Sb _0.7 Te _0.3. AgInSbTe alloys, GeSnTe alloys, and the like.
Of these, GeTe-Sb ₂ Te ₃ pseudo-binary alloys with excess Sb added, especially Ge ₁ Sb ₂ Te ₄ or Ge ₂ Sb ₂ Te ₅ near-intermetallic compounds are mainly put into practical use Has been.

  These compositions are characterized by crystallization without phase separation, which is peculiar to intermetallic compounds, and since the crystal growth rate is fast, initialization is easy, and the recrystallization rate during erasure is fast. For this reason, conventionally, a pseudo binary alloy system or a composition near an intermetallic compound has attracted attention as a recording layer showing practical overwrite characteristics (Non-Patent Document 1, Non-Patent Document 2, etc.).

However, in these compositions, metastable tetragonal crystal grains grow. The crystal grains have clear grain boundaries, irregular sizes, and remarkable optical anisotropy depending on the orientation, so that there is a problem that optical white noise is likely to occur.
And, since such crystal grains having different grain sizes and optical characteristics are likely to grow around the amorphous mark, the jitter of the mark is likely to increase, or the optical characteristics are different from the surrounding crystals. It was easy to be detected as a remaining disappearance.
For this reason, there has been a problem that good reproduction characteristics cannot be obtained in recording at a high linear velocity or high-density mark length modulation recording. Specifically, although the shortest mark length is 0.6 μm in the rewritable DVD standard, it has been found that as the shortest mark length is further reduced, the jitter increases rapidly.

Incidentally, as a measure for improving jitter, there is so-called absorption rate correction. In the conventional four-layer structure, the light energy absorbed by the recording layer is usually smaller than the light energy Aa absorbed in the amorphous state having a low reflectance (Ac). <Aa). For this reason, at the time of overwriting, there is a problem that the shape of a new recording mark changes depending on whether the original state is a crystalline state or an amorphous state, thereby increasing jitter.
This makes the absorption efficiency of light energy in the crystalline state and the amorphous state substantially the same, stabilizes the mark shape regardless of the original state, and thereby reduces jitter. Furthermore, since the crystal needs extra heat for the latent heat when melted, it is preferable that the crystal state absorbs light energy more (Ac> Aa).

  In order to achieve this relationship, there is a method in which at least one light absorbing layer is added to form five or more layers, and a part of light absorption in an amorphous state is taken away by this absorbing layer. For example, an absorption layer such as Au or Si is inserted between the lower protective layer and the substrate or on the upper protective layer (Non-patent Documents 3 and 4).

However, such a layer structure has a problem in heat resistance and adhesion of the absorbing layer, and when overwritten repeatedly, deterioration such as microscopic deformation and peeling is remarkable. Moreover, since it is easy to produce peeling etc., stability with time will also be lost.
That is, it is difficult for the GeTe—Sb ₂ Te ₃ pseudo binary alloy recording layer to achieve high density while maintaining the conventional four-layer structure.
In addition, the GeTe-Sb ₂ Te ₃ pseudo binary alloy recording layer has a wavelength dependency that the real part becomes smaller and the imaginary part becomes larger as the birefringence becomes shorter, so that a short wavelength laser beam is used as a light source. In such a case, it is difficult to achieve the condition Ac> Aa.

Therefore, in recent years, AgInSbTe quaternary alloys are being used as recording layer materials. The AgInSbTe quaternary alloy is characterized by a high erasure ratio as high as 40 dB. With the conventional four-layer structure, high-speed and high-density mark length modulation recording can be performed without correcting the absorptance. .
However, the fact that high-speed recording can be performed usually means that the crystallization speed is high and the erasure is easy, so that amorphous marks are also easily crystallized, and the recorded marks are often unstable over time.
Jpn. J. et al. Appl. Phys. , Vol. 69 (1991), p2849 SPIE, Vol. 2514 (1995), pp294-301 Jpn. J. et al. Appl. Phys. , Vol. 37 (1998), pp 3339-3342 Jpn. J. et al. Appl. Phys. , Vol. 37 (1998), pp 2516-2520

In recent years, the amount of information has increased, and in order to shorten the recording time and increase the speed of information transfer, a medium capable of recording and reproducing at a higher speed has recently been demanded. For example, the standard speed of CD (1x speed) is 1.2-1.4m / s, but CD-RW capable of recording at 4x speed has been commercialized, and recording at 8x speed and 10x speed is possible. CD-RW is required.
On the other hand, various rewritable DVDs such as DVD-RAM, DVD + RW, and DVD-RW have been proposed or commercialized. However, a 4.7 GB rewritable DVD having a capacity equivalent to that of a read-only DVD has not yet been put into practical use.
That is, there is a demand for a medium that can record a short mark at high speed and has good mark stability.

However, conventionally, high-speed recording and mark stability are considered to be contradictory properties, and it has been considered difficult to satisfy both of them simultaneously.
As a result of repeated studies on the principles of crystallization and amorphization, the present inventors have found an epoch-making medium that simultaneously satisfies all these characteristics.
That is, the present invention provides an optical information recording medium that can record short marks well at high speed and has good mark stability. More specifically, a stamper manufacturing method that can be used for the optical information recording medium is provided.

  The first gist of the present invention resides in the following stamper manufacturing method.

  That is, a stamper manufacturing method in which a phase-modulated wobble signal is applied by groove meandering, and a laser beam for photoresist exposure is applied in a radial direction according to a wobble waveform that is binary phase-modulated between ± Vw voltages. A step of exposing the photoresist layer formed on the glass master by meandering, in which the laser beam for photoresist exposure is applied to the ring modulator output wave to the EO modulator. The present invention resides in a stamper manufacturing method characterized in that

  According to the stamper manufacturing method of the present invention, disk rotation synchronization is established with reference to a wobble clock, and a data clock can be directly generated in synchronization with the wobble clock.

  In the phase change medium in which the crystalline state of the recording layer is an unrecorded / erased state and the amorphous state is a recorded state, the erasure occurs at the boundary between the amorphous part or the melted part and the peripheral crystal part. It has been found that a medium such as that formed by recrystallization which proceeds substantially can perform high-speed, high-density and stable recording. That is, it is possible to perform overwriting at high speed, to perform high-density mark length modulation recording with small mark edge jitter, and the formed mark has very good temporal stability.

In general, erasing of an amorphous mark occurs by heating the recording layer to a temperature equal to or higher than the crystallization temperature and lower than the melting point to bring it into an amorphous solid state or a molten state, and then recrystallizing it when cooled.
According to the study by the present inventors, erasing of amorphous marks, that is, recrystallization is performed by (1) crystal nucleation in an amorphous region, (2) amorphous portion or melted portion, crystal Crystal growth starting from the boundary with the part, but the former crystal nucleation is hardly caused and substantially only the latter crystal growth process is used. It was found that the following effects can be obtained.

Usually, crystallization proceeds at a crystallization temperature or more and below a melting point, but crystal nucleation proceeds at a relatively low temperature side and crystal growth proceeds at a high temperature side within that temperature range. Erasure is not impossible without crystal nucleation. Erasure is possible if crystal growth proceeds at high speed with a boundary point with a peripheral crystal region surrounding an amorphous part or a melted part as a nucleus.
In particular, a minute mark or a short mark is easily crystallized instantaneously to the center of the mark only by crystal growth from such a peripheral crystal portion, and can be completely erased in a very short time. Therefore, the effect is remarkable only in a high-density recording medium using a minute mark having a shortest mark length of 0.5 μm or less, and erasing can be performed on the order of 100 nanoseconds or less, and high-speed overwriting is possible.
In general, the shorter the mark length, the higher the recording density, but the shorter the mark length is, the more preferable is 10 nm or more from the viewpoint of the stability of the mark.

  In addition, the narrower the width of the mark, the easier it is to crystallize instantaneously to the center of the mark only by the crystal growth from the peripheral crystal part. Accordingly, the track pitch of the track on which information is recorded is preferably set to 0.8 μm or less, for example, so that the mark does not spread laterally. Usually, the mark width is about half the track pitch. In general, the narrower the track pitch, the higher the density recording is possible, but from the standpoint of the stability of the mark, 0.1 μm or more is preferable. The track may be a groove alone or both a groove and a land.

The medium of the present invention is also excellent in the stability of amorphous marks over time.
That is, crystal growth from the peripheral crystal part proceeds only in a relatively high temperature region close to the melting point and is hardly progressed at a low temperature among the crystallization temperature and the vicinity of the melting point. It is difficult to crystallize and has excellent stability over time. The crystallization temperature is usually in the range of 100 ° C. to 200 ° C., but thermal stability can be maintained up to this temperature.

In particular, in a normal use range below 100 ° C., the recorded amorphous mark is extremely stable, and the amplitude of the recorded signal hardly deteriorates. Conversely, it can be concluded that there is almost no crystal nucleation from such stability over time.
Furthermore, the medium of the present invention has an advantage that a smooth mark edge with very little fluctuation can be formed in mark length recording.
Generally, when recording an amorphous mark, the recording layer is once melted and re-solidified to make it amorphous. However, the mark edge is lower in temperature than the center. Recrystallization due to crystal nucleus growth is likely to occur, and coarse grains containing amorphous are generated, which causes mark edge fluctuations.

The medium of the present invention has a dominant and high-speed crystal growth from the boundary between the amorphous part or the melted part and the crystal part at the time of erasing. Even when the material is re-solidified and becomes amorphous, only crystal growth from the peripheral crystal part occurs, and crystallization due to crystal nucleus growth hardly occurs, and the mark edge is not easily fluctuated.
That is, crystal growth from the peripheral crystal part proceeds only in a relatively high temperature region close to the melting point and is hardly progressed at a low temperature among the crystallization temperature and the vicinity of the melting point. The boundary shape of the amorphous mark is determined only by the cooling rate at the time when the temperature drops and passes through the melting point.

Then, the coarse grains mixed with amorphous due to crystal nucleus growth that occurs at the time of resolidification, which was a problem in the past, are hardly formed around the amorphous mark. This has been found to be extremely effective in suppressing noise due to fluctuations in the mark edge.
Furthermore, since the mark edge shape is stable without changing with time, not only the initial jitter is small, but also the jitter hardly deteriorates with time.

The crystallization principle of the present invention will be described in more detail.
In this medium, the boundary between the amorphous mark and the peripheral crystal part becomes the nucleus of crystal growth, and almost no crystal nucleus is generated inside the amorphous mark. Therefore, the crystal grows only from the mark boundary.
On the other hand, in a conventional GeTe—Sb ₂ Te ₃ type recording layer, crystal nuclei are randomly generated in an amorphous mark, which grows and crystallizes.
The difference between the two crystallization processes can be confirmed with a transmission electron microscope. When both recording layers after the formation of the amorphous mark are irradiated with erasing light having a relatively low power in a direct current manner, the GeTe-Sb2Te3 recording layer is crystallized from the central part of the amorphous mark where the temperature rises. While progress is observed, in the recording layer of the present invention, it is observed that crystals grow from the periphery of the amorphous mark. In particular, crystal growth from the front end and the rear end of the amorphous mark is remarkable.

A recording layer composition that is erased by such a principle is often found in an alloy system in which excess Sb and other elements up to about 20 atomic% are added to the composition near the eutectic point of Sb _0.7 Te _0.3 . _{That, M y (Sb x Te 1} -x) 1-y 0.6 ≦ x ≦ 0.9,0 <y ≦ 0.2, M is Ga, Zn, Ge, Sn, Si, Cu, Au, Al , Pd, Pt, Pb, Cr, Co, O, S, Se, Ta, Nb, and V).
An alloy containing excess Sb in Sb _0.7 Te _0.3 has a significantly higher crystal growth from the crystal around the amorphous mark as compared with the GeTe-Sb ₂ Te ₃ pseudo binary alloy system. It has the feature that overwriting is possible. Excess Sb does not promote random crystal nucleation and crystal nucleation in the amorphous mark, but greatly increases the rate of crystal growth from the peripheral crystal part.
However, in the SbTe binary alloy, crystal nucleation occurs not a little, so the stability over time of the amorphous mark is extremely poor, and it is necessary to add an appropriate element.

According to the study by the present inventors, the addition of Ge is extremely effective in suppressing crystal nucleation.
Whether or not the recrystallization of the amorphous mark is substantially governed only by the recrystallization from the peripheral crystal part can be indirectly known from the evaluation of the temporal stability.
As a specific evaluation method, there is a method of measuring a modulation degree of a reproduction signal when an accelerated environment test under a high temperature and high humidity is performed.

That is, when a signal is recorded with a plurality of mark lengths having a minimum mark length of 0.5 μm or less, the modulation degree of the signal reproduced immediately after recording is M ₀ ,
After recording, when the modulation degree of a signal reproduced after 1000 hours at 80 ° C. and 80% RH is M ₁ ,

It is.
The mark length modulation method is not limited, and EFM modulation, EFM plus modulation, (1, 7) RLL-NRZI (run length limited-non return to zero inverted) modulation, or the like can be used. A random signal is recorded with a shortest mark length of 0.5 μm or less. In this evaluation, the shortest mark length is preferably about 0.2 μm or more. Note that it is not necessary to satisfy the above expression in all evaluation conditions, and it is sufficient to satisfy the above expression in one evaluation condition.

As an example, a random signal of an EFM plus modulation system is recorded with a plurality of mark lengths having a minimum mark length of 0.4 μm.
The modulation degree is obtained by standardizing the signal amplitude of the longest mark of the modulation method with the top signal strength. FIG. 6 shows a waveform of a DC reproduction signal (a reproduction signal including a direct current component) when a random signal subjected to EFM plus modulation is recorded and reproduced. The modulation degree is defined as the ratio I ₁₄ / I _top between the signal intensity I _{top at} the top of the 14T mark and the signal amplitude I ₁₄ .
If the degree of modulation is unchanged, it can be determined that the amorphous mark size is sufficiently stable. If the modulation degree of the random signal recorded before the acceleration test maintains 90% or more of the initial value even after the acceleration test, it can be estimated that crystal nucleation is not substantially accompanied.

  In the recording layer of the present invention, crystal growth from the peripheral crystal part is likely to occur in a high temperature region immediately below the melting point. Therefore, even when the recording layer is melted and re-solidified to form an amorphous mark, crystal growth from the peripheral crystal part Can happen. Therefore, if the cooling rate after melting is slow and does not reach the critical cooling rate required for solidification as amorphous, the entire molten region will recrystallize almost instantaneously.

This can be confirmed by the following experiment.
A (ZnS) ₈₀ (SiO ₂ ) ₂₀ first protective layer is 68 nm thick and a Ge _0.05 Sb _0.71 Te _0.24 recording layer is 18 nm thick on a 0.6 mm thick polycarbonate substrate provided with a groove for guiding recording / reproducing light. , (ZnS) ₈₀ (SiO ₂ ) ₂₀ The second protective layer has a thickness of 20 nm, Al _0.995 Ta _0.005
The reflective layer was provided with a thickness of 250 nm in this order, and an ultraviolet curable resin protective layer was provided with a thickness of 4 μm. These two sheets were laminated with a hot melt adhesive with the recording layer side facing inward to form an optical recording medium. The recording layer composition was Sb / Te≈3 so that overwriting was possible at a linear velocity of about 7 m / s or higher. This medium was initialized by scanning an elliptical laser beam having a major axis of about 100 μm and a minor axis of about 1.5 μm in the minor axis direction, melting and recrystallization.

The medium was irradiated with focused light having a wavelength of 637 nm and NA = 0.63 at a linear velocity of 7 m / s according to the guide groove. After direct recording light having a recording power Pw of 10 mW was applied in a direct current manner, the power was drastically decreased to 1 mW. That is, the recording light was substantially blocked. The beam diameter is about 0.9 μm, and corresponds to a region where the energy intensity of the Gaussian beam is 1 / e ² or more of the peak intensity.

  FIG. 2 shows the change in reflectance before and after the recording light is blocked. As shown in the lower part of FIG. 2, the recording light was cut off as time passed. The recording light is irradiated continuously, that is, in a direct current manner on the left side of the lower part of FIG. 2, and is blocked on the right side. When the same region was scanned with reproduction light having a reproduction power of 1.0 mW, a reproduction waveform as shown in the upper part of FIG. 2 was obtained. This corresponds to a change in reflectance.

  The reflectance decreases near the moment when the recording light is momentarily interrupted, and the reflectance is substantially the same before and after that. The TEM observation confirmed that the reflectivity-decreasing portion was amorphous, and was crystalline before and after that. That is, as long as the recording light is continuously irradiated, the melted portion is recrystallized, and only the melted region near the portion where the recording light is blocked becomes amorphous.

This is because when the recording light is continuously irradiated, the cooling rate of the recording layer is suppressed by the residual heat from the subsequent portion, and the critical cooling rate required for amorphous formation cannot be obtained. This is because once the light is blocked, the remaining heat from the subsequent portion can be blocked and the cooling rate can be increased.
When the recording power Pw was set to 7 mW or more, an amorphous mark was formed by blocking the recording light.

  As a result of the study, the medium of the present invention is generally recrystallized by continuously irradiating recording light having a recording power Pw sufficient to melt the recording layer at a constant linear velocity, and melts the recording layer at a constant linear velocity. It was found that an amorphous mark was formed when recording light having a recording power Pw that was sufficient to irradiate recording light having substantially zero power. The power of approximately 0 does not need to be strictly 0, but is a bias power Pb satisfying 0 ≦ Pb ≦ 0.2 Pw, more preferably a bias power Pb satisfying 0 ≦ Pb ≦ 0.1 Pw.

In the present invention, the recrystallization at the time of resolidification of the melted portion mostly occurs only by crystal growth from the surrounding solid phase crystal portion. Therefore, since the recrystallized portion is not formed at the center of the amorphous mark, a smooth and continuous mark edge is formed.
Conventionally, it has been considered that such a material that is extremely easily recrystallized is not suitable for a recording layer for mark length recording. This is because if the recording light is irradiated for a long time to form a long mark, most of the melted region is crystallized.

However, according to the study by the present inventors, in high-density recording with a shortest mark length of less than 0.5 μm, the amorphization of the molten region and recrystallization from the boundary of the surrounding solid phase crystal part Good jitter can be obtained by actively using the competitive process.
For this reason, as described later, it has been found that the pulse division method, in which the recording power Pw application section and its cutoff section, that is, the bias power Pb application section are combined, is extremely effective in forming a mark of length nT.

When recording is performed by the pulse division method, as shown in FIG. 1, an arrow mark (or crescent moon) amorphous portion is connected to form an amorphous mark.
The shape of the start end of the mark is determined only by the shape of the start end of the first arrow-shaped amorphous part, and the shape of the rear end of the mark is determined only by the shape of the rear end of the arrow-shaped amorphous part.
Usually, since the starting end shape of the amorphous part is smooth, the mark starting end shape is also smooth. This is because the cooling rate is kept sufficiently high by the escape of heat to the front, so that it almost reflects the shape of the tip of the melting region and is therefore governed by the rise time of the recording pulse. The rise of the recording pulse, that is, the Pw application section may be 2 to 3 nanoseconds or less.

On the other hand, the rear end shape of the amorphous part is determined by the cooling rate determined by the fall time of the recording pulse and the size of the recrystallized region that proceeds from the periphery, particularly the crystal part at the rear end. In order to sufficiently increase the cooling rate, the fall of the Pw application section is desirably 2 to 3 nanoseconds or less. The size of the recrystallization region can be accurately controlled by the off pulse, that is, the length of the Pb application interval.
Furthermore, by applying the above-described ultra-rapid cooling structure as the layer configuration, the cooling rate of the recording layer is made as steep as possible, and the spatial distribution of the cooling rate is made steep near the rear end of the mark so that the position of the mark end It is also important to avoid fluctuations.

As a result of intensive studies on an optical recording medium capable of recording a short mark at a high speed and having excellent recording mark aging stability, the present inventors have found that a specific composition in which Ge is added in the vicinity of the Sb _0.7 Te _0.3 eutectic composition. Was found to be particularly excellent, and by appropriately selecting the layer structure, an optical recording medium excellent in other characteristics was obtained.
That is, focusing on an unprecedented ternary alloy obtained by adding excess Sb and Ge to Sb _0.7 Te _0.3 , the suitability for high-density mark length modulation recording was examined. As a result, in the GeSbTe ternary phase diagram shown in FIG. 3, a medium using a recording layer composition with a very limited Ge—Sb—Te ratio surrounded by four straight lines A, B, C, and D In high-density mark length modulation recording, it has been found to be particularly excellent in repeated overwriting durability and stability over time.

That is, in the GeSbTe ternary phase diagram,
A straight line A connecting (Sb _0.7 Te _0.3 ) and Ge,
A straight line B connecting (Ge _0.03 Sb _0.68 Te _0.29 ) and (Sb _0.95 Ge _0.05 ),
A straight line C connecting (Sb _0.9 Ge _0.1 ) and (Te _0.9 Ge _0.1 ), and a straight line D connecting (Sb _0.8 Te _0.2 ) and Ge
A thin film mainly composed of a GeSbTe alloy having a composition of a region surrounded by the four straight lines (not including the boundary line) is used as a recording layer. By using the layer structure described later for this recording layer, the medium becomes very suitable for high-density mark length modulation recording with the shortest mark length of 0.5 μm or less. Then, a recording density equivalent to that of DVD and excellent reproduction compatibility with DVD can be obtained.

In addition, it is possible to ensure a wide margin for obtaining good jitter with respect to repeated overwriting durability and fluctuations in recording power and erasing power.
Within this composition range, the larger the amount of Sb from y = 0.7 in the Sb _y Te _1-y alloy, increasing excess amount of Sb, it is possible to overwrite at a crystallization rate is high a high linear velocity.

  More specifically, in EFM plus modulation recording (8-16 modulation mark length modulation recording), even if the length of the 3T mark, which is the shortest mark, is reduced to about 0.4 μm or 0.35 μm, good jitter is achieved. Is obtained. Also, a sufficient servo signal can be obtained, and tracking servo can be applied with an existing reproduction-only DVD drive. Furthermore, overwriting is possible at a linear velocity of 1 to 10 m / s.

As a result, a rewritable DVD having the same capacity as that of the read-only DVD and substantially compatible with reproduction can be obtained.
If the excessive amount of Sb is controlled, the above-described high-quality, high-density overwrite is possible at a high linear velocity of 8 m / s or more. Further, by changing the recording pulse dividing method (pulse strategy) according to the linear velocity as described later, it becomes possible to perform good overwriting in a wide linear velocity range including at least 3 to 8 m / s.

This composition will be described in detail below.
In the composition near Sb _0.7 Te _0.3 eutectic point where the Ge addition amount is 10 atomic% or less, the crystallization rate tends to increase as the Sb / Te ratio increases. This is because Sb in excess of Sb _0.7 Te _0.3 precipitates as Sb clusters and acts as crystal nuclei in the recrystallization process. When there is no excess Sb than Sb _0.7 Te _{0.3, the} erasing performance is insufficient and the overwriting is substantially impossible. Moreover, since there is almost no nucleation at the time of initialization, there also exists a problem that initialization is difficult and productivity is very bad (straight line A).

On the other hand, when the amount of Sb is increased in the Sb _0.7 Te _0.3 eutectic binary alloy, the crystallization temperature is lowered in exchange for an increase in the crystallization speed, and the temporal stability of the amorphous mark is impaired. . Further, it is not suitable for recording at a low linear velocity of about 3 m / s, and the formed amorphous mark disappears when irradiated with reproducing light (laser power of about 1 mW) for a short time. Therefore, Sb in excess of the straight line D connecting (Sb _0.8 Te _0.2 ) and Ge should not be included.

In addition, in the range of the excess Sb amount defined by the straight lines A and D, the SbTe binary keeps the crystallization temperature low and the presence of excess Sb crystal nuclei makes the amorphous mark unstable. Therefore, Ge is added as the excess Sb amount increases. Crystal nucleation is almost completely suppressed by the four-coordinate bond of Ge. As a result, the crystallization temperature increases and stability with time increases. A straight line B connecting (Ge _0.03 Sb _0.68 Te _0.29 ) and (Sb _0.95 Ge _0.05 ) defines this condition. More preferably, Ge is included more than the straight line B ′ connecting (Ge _0.03 Sb _0.68 Te _0.29 ) and (Sb _0.9 Ge _0.1 ).

  Further, when the Ge content is 10 atomic% or more, the jitter at the time of recording the mark length is deteriorated, and the high melting point Ge compound, particularly GeTe, is easily segregated by repeated overwriting. In addition, it is not preferable because the crystallization of the amorphous film immediately after film formation becomes extremely difficult (straight line C). In order to reduce jitter, Ge is more preferably 7.5 atomic% or less.

Note that overwriting at a linear velocity of 3m / s or more, the recording layer _{Ge x (Sb y Te 1-} y
) It is preferable to use a thin film (0.04 ≦ x <0.10, 0.72 ≦ y <0.8) mainly composed of _1-x alloy. That is, the recording at a linear velocity 3m / s or more, to increase the amount of Sb, preferably with y ≧ 0.72 in Sb _y Te _1-y alloy. However, since the stability of the amorphous mark is slightly deteriorated by increasing the amount of Sb, it is preferable to increase x ≧ 0.04 and Ge to compensate for this.
Furthermore, to overwriting at a linear velocity of 7m / s or more, the recording layer _{Ge x (Sb y Te 1-} y) 1-x alloy film on the basis of (0.045 ≦ x ≦ 0.075, 0.74 ≦ y <0.8) is preferable. That is, the recording at a linear velocity of 7m / s or more, and further increase the amount of Sb, preferably with y ≧ 0.74 in Sb _y Te _1-y alloy. At this time, the Ge amount is set to x ≧ 0.045 in order to increase the stability of the amorphous mark. On the other hand, since jitter tends to deteriorate at high linear speeds, the Ge amount is set to x ≦ 0.075 in order to compensate for this.

There have been reports on GeSbTe ternary compositions, or recording layer compositions containing additive elements based on this ternary composition (Japanese Patent Laid-Open Nos. 61-258787, 62-53886, 62). -152786, JP-A-1-63195, 1-211249, 1-2277338).
However, all of the compositions described in these are compositions that are Sb poorer than the straight line A connecting (Sb _0.7 Te _0.3 ) and Ge, and are different from the composition range of the present invention.
Rather, they are mainly based on the Sb ₂ Te ₃ metal compound composition. Further, in the GeTe-Sb ₂ Te ₃ pseudo binary alloy system, contrary to the present invention, excessive Sb has the effect of delaying the crystallization speed, and therefore, overwriting at a high linear velocity of 5 m / s or more. It is rather detrimental to include excess Sb on the GeTe—Sb ₂ Te ₃ line, especially in the Ge ₂ Sb ₂ Te ₅ composition.
As a composition in which a third element containing Ge is selectively added in the vicinity of Sb _0.7 Te _0.3 containing excessive Sb, Japanese Patent Application Laid-Open No. 1-100745 (FIG. 4 (a) composition range α), Japanese Patent Application Laid-Open No. 1-303643 No. (FIG. 4 (a) composition range β).

However, Japanese Patent Laid-Open No. 1-100745 discloses a very wide range of 0.10 ≦ x ≦ 0.80 in Sb _1-x Te _x which is a matrix composition, and uses only a region in which Sb is larger than Sb _0.7 Te _0.3. Therefore, the idea of the present application that the repeated overwriting durability and the temporal stability are excellent in high-density recording is not seen.
Japanese Patent Laid-Open No. 1-303643 does not mention the adverse effect that the stability over time of the amorphous mark is impaired if Sb exceeds the straight line D in high density recording as in the present application. In addition, none of the publications mentions the adverse effect of excessive inclusion of Ge beyond the straight line C.

Further, as the composition partially overlapping with the recording layer composition of the present invention, as shown in FIG. 4B, JP-A-1-115568 (composition range γ) and JP-A-1-251342 (composition range). δ), JP-A-3-71887 (composition range ε) and JP-A-4-28587 (composition range η).
Japanese Patent Application Laid-Open No. 1-115685 adds Au and Pd with a composition range γ as a base material, but for the purpose of low-density recording, the composition of the present invention is substantially distinguished by a straight line A and a straight line B. Yes. The composition of this publication is suitable for recording at low density corresponding to a mark length of about 1.1 μm (linear velocity 4 m / s, frequency 1.75 MHz, duty 50% square wave) and DC erasure. It is considered that a suitable composition is different from the composition of the present invention aiming at high density recording including a short mark.

The composition range δ of JP-A-1-251342 is an extremely Ge-rich GeSbTe system mainly composed of a system in which Ge is added to Sb _0.7 Te _0.3 eutectic at about 10 atomic% or more, and is linear with the composition of the present invention. It is substantially distinguished by C. In a composition containing Ge in an amount of more than 10 atomic% in the composition range δ, the crystallization speed is slow as described above, and the initialization operation for crystallizing the recording layer after film formation is difficult. There is a serious problem that it is low and cannot be put to practical use. In this publication, Au and Pd as crystal nuclei are separately added in order to overcome this problem of crystallization speed. However, in the region where Ge is smaller than the straight line C as in the present invention, such necessity is required. There is no.
Further, in this publication, it is described that if the amount of Ge is less than 10 atomic%, a sufficient change in the amount of light cannot be obtained in the recording part and the non-recording part. By devising the layer structure to be included, a very large reflected light amount change with a modulation degree of 60% or more is obtained.
The composition range ε in JP-A-3-71887 is intended for low density recording and is substantially distinguished from the composition of the present invention by a straight line C. In particular, by using the composition range of the present invention, there is no idea of the present application that it is excellent in repeated overwriting durability and stability over time in high-density recording.
The composition range η in JP-A-4-28587 includes extremely Sb-rich and Ge-rich compositions, and is substantially distinguished from the composition of the present invention by a straight line D.
As described above, none of the above publications clarifies the technical problem regarding the high-density mark length modulation recording, which is the object of the present invention, and the shortest mark length is 0.5 μm or less, For this purpose, selection of the optimum composition, improvement of the layer structure and recording method are not disclosed at all.

Next, the layer structure of the optical information recording medium of the present invention will be described. The medium of the present invention combines at least 3 m / s to 8 m / s when performing high-density mark length modulation recording with a shortest mark length of 0.5 μm or less by combining the recording layer having the above composition and the following layer structure. A medium that can be overwritten in a wide linear velocity range that preferably covers 1 m / s to 10 m / s can be realized. Then, reproduction compatibility with a so-called DVD can be maintained.
In the phase change recording layer, at least one of the upper and lower sides is covered with a protective layer.

Further, as shown in FIG. 5A, the substrate 1 / the first protective layer 2 / the recording layer 3 / the second protective layer 4 / the reflective layer 5 has a configuration, and the upper layer is made of ultraviolet or thermosetting resin. It is coated (protective coat layer 6). The order of the layers as shown in FIG. 5A is suitable when the recording layer is irradiated with a focused light beam for recording / reproduction through a transparent substrate.
Alternatively, the order of the above layers may be reversed, and a configuration in which the layers are laminated in the order of substrate 1 / reflective layer 5 / second protective layer 4 / recording layer 3 / first protective layer 2 as shown in FIG. . This layer configuration is suitable when a focused light beam is incident from the first protective layer side. Such a configuration is useful when the objective numerical aperture NA is 0.7 or more and it is necessary to reduce the distance between the recording layer and the objective lens.

If it is the structure shown to Fig.5 (a), transparent resins, such as polycarbonate, an acryl, polyolefin, or transparent glass can be used for a board | substrate.
Of these, polycarbonate resin is most preferred because it has a track record of being most widely used in CD and is inexpensive.
In the configuration shown in FIG. 5B, resin or glass can be used in the same manner, but the substrate itself does not need to be transparent. Rather, in order to increase flatness and rigidity, it is preferable to use glass or aluminum alloy. is there.
The substrate is provided with grooves having a pitch of 0.8 μm or less for guiding the recording / reproducing light. However, the grooves are not necessarily geometrically trapezoidal grooves. For example, the refractive index varies depending on ion implantation or the like. An optical groove may be formed by forming a waveguide or the like.

5A, the first protective layer 2 is provided on the substrate surface and the second protective layer 4 is provided on the recording layer 3 in order to prevent deformation due to high temperature during recording. The second protective layer 4 has a function of preventing heat from being diffused between the recording layer 3 and the reflective layer 5 and efficiently releasing heat to the reflective layer 5 while suppressing deformation of the recording layer.
Also in FIG. 5B, when viewed from the focused light beam incident side, the second protective layer 4 has functions of preventing mutual diffusion between the recording layer 3 and the reflective layer 5, heat dissipation, and deformation of the recording layer. The first protective layer in FIG. 5B has functions of preventing deformation of the recording layer, preventing direct contact between the recording layer and air (preventing oxidation contamination, etc.), and preventing damage due to direct contact with the optical pickup.

A protective layer may be further provided between the reflective layer and the substrate. For example, thermal damage to the resin substrate can be prevented.
In the configuration shown in FIG. 5B, a harder dielectric or amorphous carbon protective film or an ultraviolet or thermosetting resin layer is provided on the outer side of the first protective layer 2. Is desirable. Alternatively, a transparent thin plate having a thickness of about 0.05 to 0.6 mm can be bonded and the focused light beam can be incident through the thin plate.

Furthermore, a medium such as a DVD has a structure in which the medium shown in FIG. 5A is bonded with an adhesive with the recording layer surface inside. On the other hand, in the medium shown in FIG. Further, in the medium of FIG. 5B, a tracking groove may be formed on both surfaces of a single substrate by injection molding, and a multilayer film may be formed on both surfaces by sputtering.
The recording layer 3, the protective layers 2, 4, and the reflective layer 5 are formed by a sputtering method or the like.
In order to prevent oxidation and contamination between layers, it is desirable to form a film with an in-line apparatus in which a target for a recording layer, a target for a protective layer, and, if necessary, a target for a reflective layer material are installed in the same vacuum chamber.

The material of the protective layers 2 and 4 is determined in consideration of the refractive index, thermal conductivity, chemical stability, mechanical strength, adhesion, and the like. In general, oxides, sulfides, nitrides, carbides, and fluorides such as Ca, Mg, and Li, which are highly transparent and have a high melting point, can be used.
These oxides, sulfides, nitrides, carbides and fluorides do not necessarily have to have a stoichiometric composition, and it is also effective to control the composition for controlling the refractive index or to use them in combination. is there.

The protective layers 2 and 4 may change the composition ratio and the mixing ratio in the thickness direction. Further, each of the protective layers 2 and 4 may be composed of a plurality of films. Each film can have different materials, composition ratios, and mixing ratios according to required characteristics.
Considering the repeated recording characteristics, it is desirable from the viewpoint of mechanical strength that the film density of these protective layers is 80% or more of the bulk state. When a mixed dielectric thin film is used, the following theoretical density is used as the bulk density.

The recording layer 3 of the medium of the present invention is a phase change recording layer, and its thickness is generally preferably in the range of 5 nm to 100 nm.
If the thickness of the recording layer 3 is less than 5 nm, it is difficult to obtain a sufficient contrast, and the crystallization speed tends to be slow, so that erasing in a short time tends to be difficult.
On the other hand, if it exceeds 100 nm, it is difficult to obtain optical contrast and cracks are likely to occur.
Furthermore, it is necessary to obtain a contrast sufficient to be compatible with a read-only disk such as a DVD, and for high-density recording in which the shortest mark length is 0.5 μm or less, 5 nm to 25 nm is preferable. If the thickness is less than 5 nm, the reflectance is too low, and the influence of a non-uniform composition and a sparse film at the initial stage of film growth tends to appear, which is not preferable.

On the other hand, if it is thicker than 25 nm, the heat capacity is increased, the recording sensitivity is deteriorated, and the crystal growth is three-dimensional. Therefore, the edge of the amorphous mark tends to be disturbed and the jitter tends to increase. Furthermore, the volume change due to the phase change of the recording layer becomes remarkable, and the repeated overwrite durability deteriorates. From the viewpoint of the jitter at the mark end and the durability of repeated overwriting, it is more desirable to set it to 20 nm or less.
Further, the density of the recording layer is desirably 80% or more of the bulk density, more preferably 90% or more. As used herein, the bulk density can of course be measured by creating an alloy lump, but in the above equation (1), the molar concentration of each component is replaced with the atomic% of each element, and the bulk density is replaced by each element. An approximate value can be obtained by substituting the molecular weight of

The density of the recording layer in the sputter deposition method is such that the recording layer is irradiated by lowering the pressure of the sputtering gas (rare gas such as Ar) at the time of deposition, or by placing a substrate close to the front of the target. It is necessary to increase the amount of high energy Ar to be produced. Is high energy Ar that Ar ions irradiated to the target for sputtering partially bounce back and reach the substrate side, or whether Ar ions in the plasma are accelerated by the sheath voltage across the substrate and reach the substrate? One of them.
The irradiation effect of such a high energy noble gas is called an atomic peening effect. In sputtering with Ar gas that is generally used, Ar is mixed into the sputtered film due to an atomic peening effect. The atomic peening effect can be estimated from the amount of Ar in the film. That is, if the amount of Ar is small, it means that the effect of irradiation with high energy Ar is small, and a film with a low density is easily formed. On the other hand, if the amount of Ar is large, irradiation with high energy Ar is intense and the density becomes high, but Ar taken in the film precipitates as void during repeated overwriting and deteriorates repeated durability. An appropriate amount of Ar in the recording layer film is 0.1 atomic% or more and 1.5 atomic% or less. Further, it is preferable to use high frequency sputtering rather than direct current sputtering because the amount of Ar in the film is reduced and a high density film can be obtained.

In the present invention, the recording layer is composed of a thin film mainly composed of a GeSbTe alloy having the above composition. In other words, the ratio of the respective element amounts of Ge, Sb, and Te in the recording layer only needs to be within the above-described composition range, and other elements may be added to the recording layer up to a total of about 10 atomic% as necessary. Good.
The optical constant of the recording layer can be finely adjusted by adding at least one element selected from O, N, and S to the recording layer in an amount of 0.1 atomic% to 5 atomic%. However, it is not preferable to add more than 5 atomic% because the crystallization speed is lowered and the erasing performance is deteriorated.

Further, in order to increase the stability over time without decreasing the crystallization speed at the time of overwriting, it is preferable to add at least one of V, Nb, Ta, Cr, Co, Pt and Zr at 8 atomic% or less. . More preferably, 0.1 atomic percent or more and 5 atomic percent or less are added. The total addition amount of these additive elements and Ge with respect to SbTe is desirably 15 atomic% or less in total. If excessively contained, phase separation other than Sb is induced. In particular, when the Ge content is 3 atomic% or more and 5 atomic% or less, the effect of addition is large.
In order to improve stability over time and finely adjust the refractive index, it is preferable to add at least one of Si, Sn, and Pb at 5 atomic% or less. The total content of these additive elements and Ge is preferably 15 atomic% or less. These elements have the same 4-coordination network as Ge.

Adding Al, Ga, or In at 8 atomic% or less increases the crystallization temperature, and at the same time reduces the jitter and improves the recording sensitivity. It is preferable that Further, the content together with Ge is 15 atomic% or less, preferably 13% or less.
Addition of 8 atomic% or less of Ag is also effective in improving the recording sensitivity, and the effect is particularly remarkable when used when the Ge atomic weight exceeds 5 atomic%. However, addition of more than 8 atomic% is not preferable because it increases jitter and impairs the stability of the amorphous mark, and addition of more than 15 atomic% is not preferable because segregation is likely to occur. The most preferable Ag content is 5 atomic% or less.

The recording layer 3 of the recording medium of the present invention is usually amorphous after film formation. Therefore, after the film formation, it is necessary to crystallize the entire recording layer so as to be in an initialized state (unrecorded state).
As an initialization method, an alloy containing excess Sb in Sb _0.7 Te _0.3 can be initialized by solid-phase annealing. However, in a composition containing Ge, the recording layer is once melted and re-solidified. Initialization by melt recrystallization in which crystallization is performed by slow cooling is desirable.
This recording layer has few crystal growth nuclei immediately after film formation, and crystallization in the solid phase is difficult. However, according to melt recrystallization, a few crystal nuclei are formed and then melted. It seems that recrystallization proceeds at a high speed mainly by crystal growth.

In addition, since the recording layer of the present invention has a different reflectance from the crystal by melting recrystallization and the crystal by annealing in the solid phase, it causes noise when mixed. In actual overwrite recording, the erased portion is crystallized by melt recrystallization, and therefore initialization is preferably performed by melt recrystallization.
At this time, the recording layer is melted locally and only for a short time of about 1 millisecond or less. If the melting region is wide, or if the melting time or the cooling time is too long, each layer is destroyed by heat or the surface of the plastic substrate is deformed.
In order to provide such a thermal history, a high-power semiconductor laser beam having a wavelength of about 600 to 1000 nm is focused and irradiated on a long axis of 100 to 300 μm and a short axis of 1 to 3 μm, and the short axis direction is set as 1 scan axis. It is desirable to scan at a linear velocity of 10 m / s. Even if the same focused light is close to a circle, the melted region is too wide and re-amorphization is likely to occur, and damage to the multilayer structure and the substrate is unfavorable.
It can be confirmed as follows that the initialization was performed by melt recrystallization. That is, the recording medium with the recording power Pw focused on the spot diameter smaller than about 1.5 μm and melting the recording layer is applied to the medium after initialization at a constant linear velocity. When there is a guide groove, the tracking servo and the focus servo are applied to the track formed by the groove or between the grooves.

Thereafter, if the reflectance in the erased state obtained by irradiating the same track with the erasing light with the erasing power Pe (≦ Pw) is almost the same as the reflectance in the unrecorded initial state, the initial The crystallization state can be confirmed as a crystalline state upon melting.
This is because the recording layer is once melted by the recording light irradiation and completely recrystallized by the erasing light irradiation through the process of melting by the recording light and recrystallization by the erasing light. This is because it is in a state of being converted into a state.
Note that the reflectivity R _ini in the initialized state and the reflectivity in the melt recrystallized state R _cry are substantially the same as defined by (R _ini −R _cry ) / {(R _ini + R _cry ) / 2}. The difference in reflectance between the two is 20% or less. Usually, the reflectance difference is larger than 20% only by solid phase crystallization such as annealing.

Next, layers other than the recording layer will be described.
The layer structure of the present invention belongs to a kind of layer structure called a rapid cooling structure. The rapid cooling structure employs a layer structure that promotes heat dissipation and increases the cooling rate at the time of re-solidification of the recording layer, avoiding the problem of recrystallization when forming amorphous marks, Realize erasure ratio. For this reason, the thickness of the second protective layer is set to 5 nm or more and 30 nm or less. If the thickness is less than 5 nm, the recording layer is likely to be destroyed due to deformation or the like during melting, and the heat dissipation effect is too great, and the power required for recording becomes unnecessarily large.

The thickness of the second protective layer of the present invention greatly affects the durability in repeated overwriting, and is particularly important for suppressing the deterioration of jitter. When the film thickness is greater than 30 nm, the temperature difference between the recording side of the second protective layer and the reflective layer side becomes large during recording, and the protective layer itself is deformed asymmetrically due to the difference in thermal expansion on both sides of the protective layer. It becomes easy to do. This repetition is not preferable because it accumulates microscopic plastic deformation inside the protective layer and causes an increase in noise.
When the recording layer of the present invention is used, low jitter can be realized in high-density recording with a shortest mark length of 0.5 μm or less, but according to the study by the present inventors, a short-wavelength laser is required to realize high-density recording. In the case of using a diode (for example, a wavelength of 700 nm or less), it is necessary to pay more attention to the layer structure of the rapid cooling structure. In particular, in studying one-beam overwrite characteristics using a small focused light beam having a wavelength of 500 nm or less and a numerical aperture NA of 0.55 or more, it is possible to flatten the temperature distribution in the mark width direction to achieve a high erasure ratio and erasure. It was found that it is important for widening the power margin.

This tendency is the same in DVD-compatible optical systems using optical systems with wavelengths of 630 to 680 nm and NA = 0.6. In high-density mark length modulation recording using such an optical system, in particular, a material having low thermal conductivity is used as the second protective layer. Preferably, the film thickness is 10 nm or more and 25 nm or less.
In any case, the erasing ratio and the erasing power margin can be improved by using the reflective layer 5 provided thereon as a material having a particularly high thermal conductivity.
According to the study, in order to exhibit the good erasing characteristics of the recording layer of the present invention in a wide erasing power range, not only the temperature distribution and the time change in the film thickness direction but also the film surface direction (in the recording beam scanning direction). It is preferable to use a layer structure that can flatten the temperature distribution in the vertical direction as much as possible.

The inventors of the present invention appropriately redesigned the layer structure of the medium to flatten the temperature distribution in the track transverse direction in the medium, so that recrystallization without melting and reamorphization occurs. An attempt was made to widen the range that can be performed, and to widen the erase rate and erase power margin.
On the other hand, it was found that the temperature distribution in the recording layer was flattened by promoting heat dissipation from the recording layer to the reflective layer having an extremely high thermal conductivity through the very thin second protective layer having a low thermal conductivity. . Even if the thermal conductivity of the second protective layer is increased, the heat dissipation effect is promoted. However, if the heat dissipation is promoted too much, the irradiation power required for recording becomes high, that is, the recording sensitivity is remarkably lowered.

In the present invention, it is preferable to use a thin second protective layer having low thermal conductivity.
By using a thin second protective layer with low thermal conductivity, a time delay is given to the heat conduction from the recording layer to the reflective layer at several nsec to several tens of nsec from the start of recording power irradiation, and then the reflective layer Since heat radiation to the recording medium can be promoted, the recording sensitivity is not lowered more than necessary by heat radiation.
Conventionally known protective layer materials mainly composed of SiO ₂ , Ta ₂ O ₅ , Al ₂ O ₃ , AlN, SiN, etc. have their own thermal conductivity that is too high. The protective layer 4 is not preferable. Thus, the thermal conductivity of the metal oxide or nitride is higher by one digit or more than the following protective layer used in the protective layer of the present invention, even when compared with the same thin film state.

On the other hand, heat dissipation in the reflective layer can be achieved even by increasing the thickness of the reflective layer. However, if the thickness of the reflective layer exceeds 300 nm, the heat conduction in the film thickness direction becomes more significant than the film surface direction of the recording layer. The effect of improving the temperature distribution in the surface direction cannot be obtained. In addition, the heat capacity of the reflective layer itself increases, and it takes time to cool the reflective layer, and thus the recording layer, thereby inhibiting the formation of amorphous marks. Most preferably, a thin reflective layer having a high thermal conductivity is provided to selectively promote lateral heat dissipation.
The conventional rapid cooling structure focuses only on one-dimensional heat escape in the film thickness direction, and is intended only to release heat quickly from the recording layer to the reflective layer. Not enough attention was paid to flattening.

It should be noted that the so-called “super-quenching structure considering the heat conduction delay effect in the second protective layer” of the present invention is applied to the recording layer according to the present invention as compared with the conventional GeTe—Sb ₂ Te ₃ recording layer. More effective. This is because, in the recording layer of the present invention, the crystal growth at the time of resolidification in the vicinity of Tm is the rate-limiting factor for recrystallization. In order to increase the immediate cooling rate in the vicinity of Tm to the utmost and ensure the formation of the amorphous mark and its edge reliably and clearly, the super rapid cooling structure is effective, and the temperature distribution in the film surface direction This is because, although it was possible to erase at high speed in the vicinity of Tm by the flattening, it is possible to ensure erasure by recrystallization to a higher erasing power.

In the present invention, the material of the second protective layer is preferably low in heat conduction, but the standard is 1 × 10 ⁻³ pJ / (μm · K · nsec). However, it is difficult to directly measure the thin film thermal conductivity of such a low thermal conductivity material. Instead, a guideline can be obtained from the thermal simulation and the actual recording sensitivity measurement result.
As the second protective layer material having low thermal conductivity that brings about a preferable result, it contains at least one of ZnS, ZnO, TaS ₂ or rare earth sulfide in an amount of 50 mol% to 90 mol%, and has a melting point or decomposition point of 1000 ° C. or higher. A composite dielectric containing the above heat-resistant compound is desirable.

More specifically, a composite dielectric containing 60 mol% or more and 90 mol% or less of a rare earth sulfide such as La, Ce, Nd, or Y is desirable.
Alternatively, the composition range of ZnS, ZnO or rare earth sulfide is desirably 70 to 90 mol%.
Examples of heat-resistant compound materials having melting points or decomposition points of 1000 ° C. or higher to be mixed with these include Mg, Ca, Sr, Y, La, Ce, Ho, Er, Yb, Ti, Zr, Hf, V, Nb, Oxides such as Ta, Zn, Al, Si, Ge, and Pb, nitrides, carbides, and fluorides such as Ca, Mg, and Li can be used.
In particular, the material to be mixed with ZnO is preferably a rare earth sulfide such as Y, La, Ce, or Nd or a mixture of sulfide and oxide.
If the thickness of the second protective layer is larger than 30 nm, a sufficient flattening effect of the temperature distribution in the mark width direction cannot be obtained. Preferably it is 25 nm or less. If the thickness is less than 5 nm, the effect of delaying heat conduction in the second protective layer is insufficient, and the recording sensitivity is significantly lowered, which is not preferable.
The thickness of the second protective layer 4 is preferably 15 nm to 25 nm when the wavelength of the recording laser light is 600 to 700 nm, preferably 5 to 20 nm, more preferably 5 to 15 nm when the wavelength is 350 to 600 nm.

The present invention is characterized in that the heat dissipation effect in the lateral direction is promoted by using the thin reflective layer 5 having a very high thermal conductivity and 300 nm or less.
In general, the thermal conductivity of a thin film is significantly different from that of a bulk state, and is usually small. In particular, a thin film having a thickness of less than 40 nm is not preferable because the thermal conductivity may be reduced by an order of magnitude or more due to the influence of an island structure in the initial stage of growth. Furthermore, the crystallinity and the amount of impurities differ depending on the film forming conditions, and this causes the thermal conductivity to be different even with the same composition.

In order to define a high thermal conductivity reflective layer exhibiting good characteristics in the present invention, the thermal conductivity of the reflective layer can be directly measured, but the thermal conductivity is estimated using electrical resistance. be able to. This is because in a material such as a metal film in which electrons mainly control heat or electric conduction, the thermal conductivity and electric conductivity have a good proportional relationship.
The electric resistance of the thin film is represented by a resistivity value normalized by the film thickness or the area of the measurement region. The volume resistivity and the area resistivity can be measured by a normal four-probe method and are defined by JIS K 7194. This method provides data that is much simpler and more reproducible than actually measuring the thermal conductivity of a thin film.
In the present invention, the reflective layer preferably has a volume resistivity of 20 nΩ · m to 150 nΩ · m, more preferably 20 nΩ · m to 100 nΩ · m. A material having a volume resistivity of less than 20 nΩ · m is substantially difficult to obtain in a thin film state. Even when the volume resistivity is larger than 150 nΩ · m, the area resistivity can be lowered by using a thick film exceeding 300 nm, for example. Even if only the sheet resistivity was lowered with a rate material, a sufficient heat dissipation effect could not be obtained. This is probably because the heat capacity per unit area increases in the thick film. In addition, such a thick film is not preferable from the viewpoint of manufacturing cost because it takes time to form a film and the material cost increases. Furthermore, the microscopic flatness of the film surface is also deteriorated.
Preferably, a low volume resistivity material is used so that a sheet resistivity of 0.2 to 0.9Ω / □ can be obtained with a film thickness of 300 nm or less. 0.5Ω / □ is most preferable.

Suitable materials for the present invention are as follows.
For example, an Al—Mg—Si alloy containing 0.3 wt% to 0.8 wt% Si and 0.3 wt% to 1.2 wt% Mg.
Further, Al alloy containing Ta, Ti, Co, Cr, Si, Sc, Hf, Pd, Pt, Mg, Zr, Mo, or Mn in Al in an amount of 0.2 atomic% to 2 atomic% does not have an additive element concentration. The volume resistivity is proportionally increased, the hillock resistance is improved, and it can be used in consideration of durability, volume resistivity, film formation rate, and the like.
Regarding the Al alloy, when the amount of added impurities is less than 0.2 atomic%, the hillock resistance is often insufficient, although it depends on the film forming conditions. On the other hand, when the content is more than 2 atomic%, it is difficult to obtain the low resistivity.
When the stability over time is more important, Ta is preferable as an additive component. In particular, for the upper protective layer 4 mainly composed of ZnS, an AlTa alloy having Ta of 0.5 atomic% or more and 0.8 atomic% or less balances all of corrosion resistance, adhesion, and high thermal conductivity. Desirable as a satisfactory reflective layer. In addition, in the case of Ta, the addition of only 0.5 atomic% provides a favorable manufacturing effect in that the film formation rate during sputtering is increased by 30 to 40% compared to pure Al or Al—Mg—Si alloy.
When the Al alloy is used as the reflective layer, the preferred film thickness is 150 nm or more and 300 nm or less. If the thickness is less than 150 nm, the heat dissipation effect is insufficient even with pure Al. If it exceeds 300 nm, the heat escapes in the vertical direction from the horizontal direction and does not contribute to the improvement of the heat distribution in the horizontal direction, the heat capacity of the reflective layer itself is large, and on the contrary, the cooling rate of the recording layer becomes slow. Also, the microscopic flatness of the film surface is deteriorated.

Further, Ag, Ti, V, Ta, Nb, W, Co, Cr, Si, Ge, Sn, Sc, Hf, Pd, Rh, Au, Pt, Mg, Zr, Mo, or Mn is 0.2 atomic%. An Ag alloy containing 5 atomic% or less is also desirable. When importance is attached to stability over time, Ti and Mg are preferred as the additive component.
When the Ag alloy is used as the reflective layer, the preferred film thickness is 40 nm or more and 150 nm or less. If the thickness is less than 40 nm, the heat dissipation effect is insufficient even with pure Ag. If it exceeds 150 nm, the heat escapes in the vertical direction rather than in the horizontal direction, and does not contribute to the improvement of the heat distribution in the horizontal direction, and unnecessary thick films reduce the productivity. Also, the microscopic flatness of the film surface is deteriorated.

The present inventors have confirmed that the volume resistivity of the additive element to Al and the additive element to Ag increases in proportion to the concentration of the additive element.
By the way, it is considered that the addition of impurities generally reduces the crystal grain size, increases the electron scattering at the grain boundary, and decreases the thermal conductivity. It is necessary to adjust the amount of added impurities in order to obtain the inherent high thermal conductivity of the material by increasing the crystal grain size.
The reflective layer is usually formed by sputtering or vacuum vapor deposition, but the total impurity amount is 2 atomic%, including the amount of moisture and oxygen mixed in during film formation, in addition to the impurity amount of the target and vapor deposition material itself. It is necessary to do the following. Therefore, it is desirable that the ultimate vacuum of the process chamber is 1 × 10 ⁻³ Pa or less.
Further, when forming a film at an ultimate vacuum lower than 10 ⁻⁴ Pa, it is desirable to prevent the incorporation of impurities by setting the film formation rate to 1 nm / second or more, preferably 10 nm / second or more.

Alternatively, when the intentional additive element is included in an amount of more than 1 atomic%, it is desirable to prevent the addition of additional impurities as much as possible by setting the film formation rate to 10 nm / second or more.
The film formation conditions may affect the crystal grain size regardless of the amount of impurities. For example, in an alloy film in which about 2 atomic% of Ta is mixed in Al, an amorphous phase is mixed between crystal grains, but the ratio between the crystalline phase and the amorphous phase depends on the film forming conditions. For example, as the sputtering is performed at a lower pressure, the proportion of the crystal portion increases, the volume resistivity decreases, and the thermal conductivity increases.
The impurity composition or crystallinity in the film also depends on the manufacturing method of the alloy target used for sputtering and the sputtering gas (Ar, Ne, Xe, etc.).
Thus, the volume resistivity in the thin film state is not determined only by the metal material and composition.
In order to obtain high thermal conductivity, it is desirable to reduce the amount of impurities as described above. On the other hand, pure metals such as Al and Ag tend to be inferior in corrosion resistance and hillock resistance. The optimum composition is determined in consideration.

In order to obtain higher heat conduction and higher reliability, it is also effective to make the reflective layer multilayer. At this time, at least one layer substantially controls the heat dissipation effect as the low volume resistivity material having a thickness of 50% or more of the total reflection layer thickness, and the other layers have corrosion resistance, adhesion to the protective layer, and resistance to resistance. It is configured to contribute to improving hillocks.
More specifically, Ag, which has the highest thermal conductivity and low volume resistivity among metals, has a poor compatibility with the protective layer containing S, and tends to deteriorate somewhat when repeatedly overwritten.
In addition, corrosion tends to occur in an accelerated test environment of high temperature and high humidity.
Therefore, it is also effective to use Ag and an Ag alloy as the low volume resistivity material, and to provide an alloy layer containing Al as a main component as an interface layer between the upper protective layer and the upper protective layer in the range of 1 nm to 100 nm. If the thickness is 5 nm or more, the layer does not have an island-like structure and is easily formed uniformly.
As the Al alloy, as described above, for example, an Al alloy containing Ta, Ti, Co, Cr, Si, Sc, Hf, Pd, Pt, Mg, Zr, Mo, or Mn in an amount of 0.2 atomic% to 2 atomic%. Is mentioned.
If the thickness of the interface layer is less than 1 nm, the protective effect is insufficient, and if it exceeds 100 nm, the heat dissipation effect is sacrificed.
The use of the interface layer is particularly effective when the reflective layer is Ag or an Ag alloy. This is because Ag is relatively easily corroded by sulfuration due to contact with a protective layer containing sulfide, which is preferable in the present invention.

Further, when an Ag alloy reflective layer and an Al alloy interface layer are used, since Ag and Al are relatively easy to interdiffuse, it is more preferable to provide an interface oxide layer by oxidizing the Al surface to be thicker than 1 nm. If the interfacial oxide layer exceeds 5 nm, particularly 10 nm, it becomes a thermal resistance, which is not preferable because the function as a reflective layer with extremely high heat dissipation, which is the original purpose, is impaired.
The multilayer reflection layer is also effective for obtaining a desired area resistivity with a desired film thickness by combining a high volume resistivity material and a low volume resistivity material.
The volume resistivity adjustment by alloying can simplify the sputtering process by using an alloy target, but it also increases the target manufacturing cost and consequently the raw material ratio of the medium. Therefore, it is also effective to obtain a desired volume resistivity by multilayering a thin film of pure Al or pure Ag and a thin film of the additive element itself.
If the number of layers is up to about three layers, the initial apparatus cost increases, but the individual medium costs may be suppressed.
The reflective layer is a multilayer reflective layer made of a plurality of metal films, the total film thickness is 40 nm or more and 300 nm or less, and a metal thin film layer having a volume resistivity of 20 nΩ · m or more and 150 nΩ · m or less is 50% or more of the thickness of the multilayer reflection layer (It may be a multilayer).
The thickness of the recording layer and the protective layer has good laser light absorption efficiency in consideration of the interference effect associated with the multilayer structure in addition to the limitations on the thermal characteristics, mechanical strength, and reliability. The signal amplitude, that is, the contrast between the recorded state and the unrecorded state is selected to be large.

For example, if the medium of the present invention is applied to a rewritable DVD to ensure compatibility with a read-only DVD, the modulation degree must be high. In addition, it is necessary to be able to apply a tracking servo method called a DPD (Differential Phase Detection) method, which is usually used in a reproduction-only player, as it is.
FIG. 6 shows a waveform of a DC reproduction signal (a reproduction signal including a direct current component) when a random signal subjected to EFM plus modulation is recorded and reproduced. The modulation degree is defined as the ratio I ₁₄ / I _top between the signal intensity I _{top at} the top of the 14T mark and the signal amplitude I ₁₄ .
I _top actually corresponds to the reflectance in the groove of the unrecorded portion (crystalline state). I ₁₄ has a problem of intensity difference and phase difference of reflected light from the crystal part and the amorphous part of the phase change medium.
The difference in the intensity of the reflected light is basically determined by the difference in reflectance between the crystalline state and the amorphous state. If the modulation degree after recording is approximately 0.5 or more, low jitter can be realized, and the tracking servo by the DPD method also operates well.

FIG. 7 shows a calculation example of the reflectance difference in a typical four-layer configuration. (ZnS) ₈₀ (SiO ₂ ) ₂₀ protective layer, Ge _0.05 Sb _0.69 Te _0.26 recording layer on a polycarbonate substrate, (ZnS
) ₈₀ (SiO ₂ ) ₂₀ protective layer, Al _0.995 Ta _0.005 reflective layer provided.
The measured value is used for the refractive index of each layer. The complex refractive index of each material at a wavelength of 650 nm is as follows. The upper and lower protective layers are 2.12-0.0i, the reflective layer is 1.7-5.3i, the substrate is 1.56, and the recording layer is in an amorphous state. 3.5-2.6i in the state immediately after the film) and 2.3-4.1i in the crystal state after initialization.
Further, the film thicknesses of the recording layer, the second protective layer, and the reflective layer were fixed at 18 nm, 20 nm, and 200 nm, respectively.
As far as the dependence on the thickness of the first protective layer is seen, the change in amplitude is usually small and strongly depends on the denominator I _top , that is, the reflectivity of the crystalline state. Therefore, it is desirable that the crystal state reflectance be as low as possible.

In the calculation example of FIG. 7, the first protective layer is made of (ZnS) ₈₀ (SiO ₂ ) ₂₀ having a refractive index n = 2.12.
A membrane was obtained. At this time, the first minimum value d ₁ has a thickness of 50 to 70 nm, and the second minimum value d ₂ has a thickness of 200 to 220 nm. Thereafter, it changes periodically.
The film thickness of the first protective layer that minimizes the reflectance in the crystalline state is substantially determined only by the refractive index of the protective layer in the case of a recording layer having a high reflectance. The minimum point film thickness at another refractive index n can be obtained by multiplying d ₁ and d ₂ by 2.1 / n, but normally, a dielectric used as a protective layer is about n = 1.8 to 2.3. D ₁ is about 60 to 80 nm.
If the refractive index n of the first protective layer is smaller than 1.8, the reflectance at the minimum point increases, the modulation degree is significantly lowered, and is less than 0.5, which is not preferable. On the other hand, if the value is 2.3 or more, the reflectance of the minimum point becomes too low to achieve 20%, which is not preferable because focus and tracking servo become difficult.

In the composition range of the recording layer according to the present invention, optical characteristics almost similar to those in FIG. 7 are exhibited.
From the viewpoint of productivity, it is desirable that the thickness of the first protective layer be 150 nm or less. This is because, at present, the deposition rate of the dielectric protective layer by the sputtering method is at most 15 nm / second, and it takes 10 seconds or more for the deposition to increase the cost. Moreover, since the allowable value of film thickness fluctuation becomes strict, it is not preferable in production. That is, as can be seen from FIG. 7, when the reflectance deviates from the desired film thickness d ₀ by Δd, the reflectance fluctuates in the same manner both in the vicinity of the first minimum value d _{1 and in the} vicinity of the second minimum value d ₂ .

On the other hand, the film thickness distribution of the production is usually a limit 2 to 3% ± the uniformity with respect to d _0. Therefore, the thinner the d ₀ , the smaller the fluctuation range Δd of the film thickness, which is advantageous because the fluctuation of the reflectance within the disk surface or between the disks can be suppressed.
Therefore, it is desirable to adopt a film thickness in the vicinity of the first minimum value d _{1 in} an inexpensive stationary facing type sputtering apparatus that does not have a substrate revolving mechanism.
On the other hand, since a thick protective layer has a great effect of suppressing deformation of the substrate surface during repeated overwriting, a film thickness in the vicinity of the second minimum value d ₂ is adopted if importance is attached to improving the durability of repeated overwriting. It is desirable to do.
In a medium in which recording / reproducing light is incident through a substrate for recording or reproduction, the first protective layer must be thickened to some extent to protect the substrate from heat generated during recording. At the time of recording, the recording layer is about 100 nanoseconds but is 500 to 600 ° C. or higher. For this purpose, the film thickness is preferably 50 nm or more. If the thickness is less than 50 nm, microscopic deformation is accumulated on the substrate when recording is repeated, and noise and defects are likely to occur. This is particularly important when the substrate is made of a thermoplastic such as polycarbonate.

Next, a preferred optical recording method for use with the present medium will be described.
A preferred first recording method is to record information on which the mark length has been modulated with a plurality of recording mark lengths on the recording medium described above.
Between the recording marks, irradiating recording light with an erasing power Pe that can crystallize amorphous,
When the time length of one recording mark is nT (T is a reference clock period, n is an integer of 2 or more),
The time length nT of the recording mark is

(Where m is the number of pulse divisions, m = n−k, and k is an integer satisfying 0 ≦ k ≦ 2.
Σ _i (α _i + β _i ) + η ₁ + η ₂ = n, η ₁ is a real number satisfying η ₁ ≧ 0, η ₂ is a real number satisfying η ₂ ≧ 0, and 0 ≦ η ₁ + η ₂ ≦ 2.0 .
α _i (1 ≦ i ≦ m) is a real number such that α _i > 0, β _i (1 ≦ i ≦ m) is a real number such that β _i > 0, and Σα _i <0.5n.
α ₁ = 0.1 to 1.5, β ₁ = 0.3 to 1.0, β _m = 0 to 1.5, α _i = 0.1 to 0.8 (2 ≦ i ≦ m) To do.
Note that α _i + β _i-1 = 0.5 to 1.5 in i where 3 ≦ i ≦ m, and is constant regardless of i. )
Divided in the order of
Within the time of α _i T (1 ≦ i ≦ m), the recording light Pw of Pw ≧ Pe for melting the recording layer is irradiated, and within the time of β _i T (1 ≦ i ≦ m). Is 0 <Pb ≦
Recording light with a bias power Pb of 0.2 Pe (however, in β _m T, 0 <Pb ≦ Pe can be satisfied) is irradiated.

By using this recording method in combination with the above-mentioned medium, the cooling rate at the time of re-solidification of the recording layer can be accurately controlled, and at least 3 m / s to 8 m / s, and further, 1 m / s to 15 m depending on the setting of recording conditions. High-density mark length modulation recording with a shortest mark length of 0.5 μm or less is possible in a wide linear velocity range of / s, a repeat overwriting of 1000 times or more can be achieved, and a low jitter of less than 10% of the reference clock period T realizable.
First, in order to realize the high-density mark length modulation recording as described above, a laser beam having a wavelength of 350 to 680 nm is condensed on the recording layer through an objective lens having a numerical aperture NA of 0.55 to 0.9. To obtain a fine focused light beam spot.

More preferably, NA is 0.55 or more and 0.65 or less. If the NA exceeds 0.65, the influence of aberration due to the tilt of the optical axis becomes large, so it is necessary to make the distance between the objective lens and the recording surface very close. Therefore, when the focused light beam is incident through a substrate having a thickness of about 0.6 mm such as a DVD, the upper limit of NA is about 0.65.
As shown in FIG. 8, the power margin and the recording linear velocity margin can be widened by modulating the recording light power to at least three values.
In FIG. 8, the start position of the _first recording pulse α ₁ T and the end position of the final off pulse β _m T do not necessarily coincide with the start position and end position of the original recording signal. Within the range of 0 ≦ η ₁ + η ₂ ≦ 2.0, η ₁ T may be placed at the beginning and η ₂ T may be placed at the end. Fine adjustment of the lengths of η ₁ T and η ₂ T according to the length of the mark before and after the mark and the length between marks is also effective in accurately forming the mark.

Alternatively, a good mark may be formed by changing only β _m according to the mark length nT. The last β _m = 0 may be set. For example, 3T-1 in EFM modulation
11T mark among the marks 1T, or 14T mark among the marks 3T~14T in EFM plus modulation, the heat tends to accumulate longer mark the like, to the longer the cooling time by increasing the final beta _m Is good.
Conversely, in the case of a short mark such as a 3T mark, β _m should be shortened. The adjustment range is 0
. About 5. In the case of high-density recording that exceeds the linear recording density of a so-called DVD, sufficient recording signal quality can be obtained without necessarily performing such fine adjustment.

The mark shape can also be controlled by changing the magnitude of the bias power Pb. FIG. 9 shows an example of temporal change in temperature at one point of the recording layer when two recording pulses are irradiated. This is a temperature change at the position where the recording pulse P1 is irradiated when the recording pulse P1, the off pulse, and the recording pulse P2 are continuously irradiated while moving the beam relative to the medium. (A) is when Pb = Pe, and (b) is when Pb≈0.
In FIG. 9B, since the bias power Pb in the off-pulse interval is almost 0, TL ′ falls to a point sufficiently lower than the melting point, and the cooling rate in the middle is also high. Therefore, the amorphous mark is melted at the time of irradiation with the recording pulse P1, and formed by rapid cooling at the subsequent off-pulse.
On the other hand, in FIG. 9A, since the erasing power Pe is irradiated even in the off-pulse period, the cooling rate after the first recording pulse P1 irradiation is slow, and the lowest temperature TL reached by the temperature drop in the off-pulse period is the melting point Tm. It remains in the vicinity and is further heated to the vicinity of the melting point Tm by the subsequent recording pulse P2, and it is difficult to form an amorphous mark.

Taking a steep temperature profile as shown in FIG. 9B for the medium of the present invention is important in suppressing crystallization in a high temperature range and obtaining a good amorphous mark. is there. This is because the recording layer of the medium of the present invention exhibits a high crystallization rate only in a high temperature region just below the melting point, and the recrystallization is suppressed by taking the profile (b) where the recording layer temperature hardly remains in the high temperature region. This is because it is considered possible.
Alternatively, crystal nucleation in a relatively low temperature region close to the crystallization temperature Tc is not dominant in each erasing process, and Sb clusters that can be crystal nuclei formed at the time of the above-described initialization exist stably. It is considered that only the crystal growth in the region is dominant.
Therefore, by controlling the cooling rate and TL ′, recrystallization is almost completely suppressed, and an amorphous mark having a clear outline substantially coinciding with the melted region can be obtained, and jitter at the mark edge can be reduced.

On the other hand, in the GeTe—Sb ₂ Te ₃ pseudo binary alloy, there is no significant difference in the amorphous mark formation process in any of the temperature profiles of FIGS. 9A and 9B. This is because this material shows recrystallization although the speed is slightly low even in a wide temperature range, particularly in a low temperature region near the crystallization temperature Tc. Alternatively, in this material, crystal nucleation in a temperature range relatively close to Tc and crystal growth in a temperature range close to Tm are rate-determined, so that relatively low-speed recrystallization occurs in a wide temperature range as a whole. It is thought that happens.
Even in GeTe-Sb ₂ Te ₃ , coarse grains may be suppressed by using an off pulse as Pb <Pe. However, when Pb / Pe ≦ 0.2, crystallization in the vicinity of Tc is excessively suppressed. Erase performance decreases.
However, in the recording layer material according to the present invention, it is considered that crystallization at a relatively low temperature close to Tc hardly proceeds, and therefore, it is preferable that Pb / Pe ≦ 0.2. Or more specifically, when 0 ≦ Pb ≦ 1.5 (mW), it is preferable to use as low a Pb as possible so that the tracking servo is stable, and to actively use an off pulse so as to be cooled as quickly as possible. The edge can be clearly formed, which is preferable.

In the pulse division method of FIG. 8, in particular, only the most advanced recording pulse α ₁ T is made longer than the succeeding pulse α _i T, and only the leading and trailing off pulse widths β ₁ T and β _m T are set to other values. Setting this separately from the off pulse is most effective in balancing the characteristics of the long mark and the short mark.
The state-of-the-art pulse α ₁ T does not have a residual heat effect, and therefore requires a little longer time for temperature increase. Alternatively, it is also effective to set the recording power of the most advanced pulse higher than the subsequent pulses.

Further, if the pulse switching is synchronized with the clock cycle T, the pulse control is simplified. FIG. 10 shows a pulse division method suitable for mark length modulation recording and having a simple pulse control circuit. As a pulse division method for recording the mark length modulation data of (a), (b) shows the case of m = n-1, and (c) shows the case of m = n-2. In (b) and (c), T is omitted for the sake of simplicity. In either case, α _i (2 ≦ i ≦ m) and β _i (2 ≦ i ≦ m−1) are constant regardless of i, and α ₁ ≧ α _i , α _i + β _i-1 = 1.0 (3 ≦ i ≦ m), α _i (2 ≦ i ≦ m)
The trailing edge of the recording pulse is synchronized with the clock pulse.
It is also effective for simplifying the circuit to make Pb the same as the reproduction light power Pr. Making only the _first pulse α ₁ T longer than the succeeding pulse is necessary to improve the recording balance between the short mark and the long mark in the so-called eye pattern. Alternatively, only the head pulse may have higher power than the subsequent pulse.
Such a pulse can be achieved by determining three types of gate generation circuits as shown in FIG.

FIG. 11 is an explanatory diagram of an example of a pulse generation method according to the recording method of the present invention. (A) is a clock signal, (b) is a data signal, which are gate signals (c) Gate1, (d) Gate2, and (e) Gate3 generated from three types of gate generation circuits in the recording pulse generation circuit. By determining the priority order of these three types of gate signals, the pulse division method of the present invention can be achieved.
Gate _{1 determines} only the recording pulse generation interval α ₁ T, and Gate 2 determines the timing for generating a predetermined number of subsequent pulses α _i T (2 ≦ i ≦ m). Here, the pulse width α _i is set to a constant value α _c when 2 ≦ i ≦ m. Gate 3 generates an off-pulse generation period β _i T. While Gate 3 is on (level high), Pb is generated, and when it is off (level low), Pe is generated.
By determining only the rising timing of α ₁ and the pulse width independently, β ₁ can be set to a value different from β _i .
The rise of Gate3 and Gate1 should be synchronized. Gate1 and Gate2 each generate Pw, but when Gate1 and Gate2 are on, they have priority over Gate3. If the delay times T ₁ and α ₁ of Gate ₁ and the delay times (T ₁ + T ₂ ) and α _c of Gate ₂ are designated, the strategy of FIG. 10 can be designated.

Here, if T ₁ is 1T or more, the pulse in the case of m = n−1 in FIG.
If it is less than 1T and the number of subsequent pulses is reduced by one, it becomes a pulse in the case of m = n−2 in FIG. At this time, the mark length to be formed is set to nT by making α ₁ T and β _m−2 T longer than in the case of m = n−1.
As a further application example of the present invention, it is desirable to use the following recording method in order to obtain a signal quality equivalent to or higher than that of a read-only DVD at a recording density equivalent to or higher than that of a read-only DVD.

That is, an optical recording method for recording and reproducing data by condensing light having a wavelength of 350 to 680 nm on a recording layer through an objective lens having a numerical aperture NA of 0.55 to 0.9, and m = n− 1 or m = n−2, 0 ≦ Pb ≦ 1.5 (mW), and Pe / Pw is 0.3 or more and 0.6 or less. And
α ₁ = 0.3 to 1.5,
α ₁ ≧ α _i = 0.2 to 0.8 (2 ≦ i ≦ m),
α _i + β _i-1 = 1.0 (3 ≦ i ≦ m),
It is preferable that β _m = 0 to 1.5.

  The reason why the ratio of Pe / Pw is kept constant is to increase the erasable range by increasing the erasing power when the power fluctuation occurs and the recording mark is large at a high power. If Pe / Pw is less than 0.3, Pe is always low and erasure is apt to be insufficient. On the other hand, if it is larger than 0.6, Pe is excessive and re-amorphization at the center of the beam tends to occur, and erasure by complete recrystallization becomes difficult. In addition, the amount of energy applied to the recording layer becomes too large, and the recording layer tends to deteriorate due to repeated overwriting.

In the recording layer having the composition according to the present invention, good jitter can be obtained in a range where α _i is particularly small. Therefore, Σα _i <0.5n, and (Σα _i ) / n can be reduced as k is decreased. desirable. That is, it is preferable that (Σα _i ) <0.4n when k = 0 or k = 1, and (Σα _i ) <0.5n when k = 2.
Preferably, in order to apply such a recording pulse division method to overwriting at a linear velocity 3m / s or more, in the present invention the recording layer _{Ge x (Sb y Te 1-} y) 1-x, in particular y For overwriting at 0.72 or more and a linear velocity of 7 m / s or more, y is set to 0.74 or more. That is, the Sb / Te ratio is 2.57 or more, more preferably 2.85 or more.

In the present invention, even if the recording layer composition is Sb rich in this way, it is one of the preferable features that the stability of the amorphous mark is high and the storage stability is also good.
Japanese Patent Application Laid-Open No. 8-22644 describes an AgInSbTe recording layer in which Ag and In are added to a composition in the vicinity of Sb _0.7 Te _0.3 in a total amount of about 10 atomic%. However, when the Sb / Te ratio is 2.57 or more in this AgInSbTe recording layer, the amorphous mark becomes extremely unstable and there is a problem in storage stability.
Hereinafter, a comparative description will be given using experimental examples. When performing EFM plus modulation mark length recording, an optical system having a wavelength of 630 to 680 nm and NA of 0.6 is used in order to record a mark having a length of nT in a linear velocity range of 2 m / s to 5 m / s. Consider a case where recording is performed by dividing the recording pulse into n-1.
As an example of the recording layer of the present invention, Ag _0.05 Ge _0.05 Sb _0.67 Te _0.23 (Sb / Te≈2.91) is used, and as an example of the AgInSbTe recording layer, Ag _0.05 In _0.05 Sb _0.63 Te _0.27 (Sb / Te≈2). .33) is used.

Since the optical constants of the recording layer of the composition of the present invention and the above AgInSbTe recording layer are almost the same, it is possible to obtain the same reflectivity and degree of modulation using the same layer configuration, and therefore a thermally equivalent layer configuration. Applicable.
The film thickness of the first protective layer is 100 nm, the recording layer is 20 nm, the second protective layer is 20 nm, and the reflective layer is 200 nm. In each case, β _i = about 0.5 (1 ≦ i ≦ n−1), Pw = 10 14 (mW), Pe / Pw = 0.5, and Pb≈0.
At this time, in the conventional Ag _0.05 In _0.05 Sb _0.63 Te _0.27 recording layer, α ₁ = 0.8 to ₁ .
2, α _i = 0.4 to 0.6 (2 ≦ i ≦ n−1) is preferable. In particular, α ₁ = 1.0, α _i (2 ≦
When i ≦ n−1) = 0.5 and β _m = 0.5, Σα _i is 0.5n regardless of n.

On the other hand, in the Ag _0.05 Ge _0.05 Sb _0.67 Te _0.23 recording layer of the present invention, α ₁ = 0.3 to 0.5.
, Α _i = 0.2 to 0.4 (2 ≦ i ≦ n−1) is a preferable range. More specifically, α ₁ = 0.6 and α _i (2 ≦ i ≦ n−1) = 0.35. In this case, when n = 3, the Σα _i ≒ 0.32n next, n = 4 or more, and Σα _i ≒ 0.33n~0.34n.
This means that in the medium of the present invention, the average irradiation power irradiated during recording can be reduced and the substantial recording pulse irradiation time can be reduced to Σα _i <0.4n.

As a result, the following effects can be obtained.
(1) Deterioration of recording signal quality due to high power recording can be reduced. The problem of high power recording is due to the fact that too much light energy is applied to the recording layer and the recording layer is trapped. For this reason, the cooling rate becomes slow, recrystallization of the amorphous mark occurs, and deterioration during repeated overwriting becomes remarkable.
By providing a low-power off-pulse section, the average input power is suppressed, and heat is released in the plane direction by a reflective layer with high thermal conductivity, so even at high power recording, the trailing edge of the mark, especially the trailing edge of the long mark The adverse effect of heat accumulation can be suppressed, and a good long mark can be formed.
(2) The thermal damage of each layer during repeated overwriting can be reduced, and repeated durability can be improved. By reducing the thermal damage each time, for example, deformation of a plastic substrate that is vulnerable to heat can be suppressed. Further, the range of damage can be limited to a narrower range at the center of the laser beam profile.
In particular, the effect of reducing the actual recording energy irradiation ratio (Σα _i ) / n is greater for n = 4 or more long marks where heat is easily stored. Therefore, it is easily damaged by heat 5m / s
Even at the following low linear velocity, adverse effects on the medium can be reduced.

In the present invention, the repeated overwriting durability can be improved as described above, and the number of overwriting times one digit or more larger than the conventional one can be achieved.
Furthermore, the recording layer, a _{Ge x (Sb y Te 1-} y) 1-x alloy thin film mainly (0.045 ≦ x ≦ 0.075,0.74 ≦ y <0.8), the line By making the recording pulse dividing method variable in accordance with the speed, overwriting can be performed in a wide range of linear velocities including 3 m / s to 8 m / s.
That is, in the pulse division method of FIG. 8, k of m = n−k is constant, and either Pb / Pe or α _i is monotonously decreased as the linear velocity at the time of overwriting is lower.
In order to keep the recording linear density constant, the clock cycle may be changed according to the linear velocity, or the Pw and Pe may be changed to be optimally maintained at the respective linear velocities as needed. .

The present invention further provides a method for recording a so-called EFM plus modulation signal in which the shortest mark length is 0.35 to 0.45 μm at both 1 × and 2 × speeds of the standard reproduction linear velocity of DVD. Note that the standard reproduction linear velocity of DVD is 3.49 m / s.
That is, light having a wavelength of 600 to 680 nm passes through an objective lens having a numerical aperture NA of 0.55 to 0.65 and is condensed on the recording layer through the substrate, and the shortest mark length is 0.35 to 0.45 μm. An optical recording method for recording and reproducing data,
n is an integer from 1 to 14,
m = n−1,
Pb is constant in the range of 0 ≦ Pb ≦ 1.5 (mW) regardless of the linear velocity,
Pe / Pw can be changed in accordance with the linear velocity in the range of 0.4 to 0.6. (I) In the range of the recording linear velocity of 3 to 4 m / s, the reference clock cycle is To.
α ₁ = 0.3 to 0.8,
α ₁ ≧ α _i = 0.2 to 0.4 and constant regardless of i (2 ≦ i ≦ m),
α ₂ + β ₁ ≧ 1.0,
α _i + β _i-1 = 1.0 (3 ≦ i ≦ m),
β _m = 0.3 to 1.5,
In the time of α _i T (1 ≦ i ≦ m), the recording light of the recording power Pw ₁ is irradiated. (ii) In the range of the recording linear velocity of 6 to 8 m / s, the reference clock cycle is To / 2. ,
α ′ ₁ = 0.3 to 0.8,
α ′ ₁ ≧ α ′ _i = 0.3 to 0.5 and constant regardless of i (2 ≦ i ≦ m),
α ′ _i + β ′ _i−1 = 1.0 (3 ≦ i ≦ m),
β ′ _m = 0 to 1.0,
When the recording light of recording power Pw ₂ is irradiated within the time of α _i T (1 ≦ i ≦ m),
α ′ _i > α _i (2 ≦ i ≦ m),
In this optical recording method, 0.8 ≦ Pw ₁ / Pw ₂ ≦ 1.2. According to the experiments by the present inventors, particularly good jitter was obtained with this setting as long as the pulse division method of FIG. 10 was used.

Here, if α ₂ + β ₁ = 1.0, the independent parameter related to the pulse width is α ₁ ,
The number of α _i and β _m is three, which is preferable because the recording signal source can be further simplified.
In addition, it is not necessary to take all the integers from 1 to 14 as n, and 3 to 11 and 14 are taken in EFM plus modulation. A (1, 7) RLL-NRZI (Run Length Limited-Non Return To Zero Inverted) code or the like can also be used.
In order to keep the recording density constant, generally, the clock cycle during 1 × speed recording is set to be double that during 2 × speed recording.
In the present invention, recording is performed not only on the entire recording area while maintaining a constant linear velocity (constant linear velocity, CLV method) as described above, but also on the entire recording area at a constant rotational angular velocity. It is also effective for the method (constant angular velocity, CAV method). Alternatively, it is also effective for a ZCLV (Zoneed CLV) system in which the radial direction is divided into a plurality of zones and overwriting is performed by the CLV system in the same zone.
The diameter of the optical disk varies, such as 86 mm, 90 mm (single CD size), 120 mm (CD size), or 130 mm, and the recording area ranges from a radius of 20-25 mm to a maximum of 65 mm. At this time, the linear velocity difference between the inner and outer circumferences is nearly three times the maximum.

In general, in high-density mark length recording, the linear velocity range in which the phase change medium exhibits good overwrite characteristics is a range of about 1.5 times the linear velocity ratio. If the linear velocity is high, the cooling rate of the recording layer is high, so that amorphous marks are likely to be formed, but the time that is maintained above the crystallization temperature is shortened and erasure becomes difficult. On the other hand, if the linear velocity is slow, erasure is easy, but the cooling rate of the recording layer is slow, so that recrystallization is likely to occur and a good amorphous mark is difficult to be formed.
In order to solve this problem, it is possible to adjust the thickness of the reflective layer on the inner and outer peripheries to increase the heat radiation effect by the reflective layer on the inner perimeter. Alternatively, it has been proposed to change the recording layer composition to increase the crystallization rate at the outer periphery, or to lower the critical cooling rate necessary for amorphous formation at the inner periphery. However, it is not easy to create a disk having such a distribution.

On the other hand, according to the combination of the medium of the present invention and the optical recording method, if the linear velocity at the outermost circumference of the disk, that is, the maximum linear velocity is about 10 m / s or less, good recording is possible even in the CAV method and ZCLV method. Is possible.
In order to use the present invention for a medium whose linear velocity changes according to the radius as described above, the recording area is divided into a plurality of zones by the radius, and the reference clock frequency and the pulse division method of the data are switched for each zone. It is desirable to use.

That is, a method for recording information with a plurality of mark lengths by rotating an optical information recording medium having a predetermined recording area at a constant angular velocity, wherein the linear velocity at the innermost circumference of the recording area is 2 to 4 m / s. The medium is rotated so that the linear velocity at the outermost periphery of the recording area is 6 to 10 m / s. The recording area is composed of a plurality of zones divided by a radius, and recording is performed according to the average linear velocity in each zone. The reference clock period T is changed so that the density is substantially constant.
At this time, the pulse division number m is made constant regardless of the zone, and the Pb / Pe ratio and / or α _i (where i is at least one of 1 ≦ i ≦ m) monotonically from the outer peripheral zone toward the inner peripheral zone. Decrease. As a result, it is possible to prevent incomplete formation of the amorphous mark due to an insufficient cooling rate in the inner periphery of the low linear velocity. Incidentally, alpha _i (i is at least one of 1 ≦ i ≦ m) reduces monotonously, and, for example alpha _1, alpha _2, · · ·, to decrease only alpha ₂ in the alpha _m Point to.
More specifically, it is desirable to use the pulse division method according to the linear velocity based on the pulse division method shown in FIG. 10 because the variable pulse division method circuit can be simplified. At this time, dividing the recording area into p zones in the radial direction and changing the clock period and pulse dividing method for each zone is easier than continuously changing according to the radial position. .

In the present invention, the recording area is divided into p zones by the radius, the innermost circumferential side is the first zone, the outermost circumferential side is the pth zone, and the qth zone (where q is an integer of 1 ≦ q ≦ p) ) Is the angular velocity ω _q , the average linear velocity is <v _q > _ave , the maximum linear velocity is <v _q > _max , the minimum linear velocity is <v _q > _min , the reference clock cycle is T _q , and the time length of the shortest mark Let n _min T _q be
_{<V p> ave / <v} 1> ave is in the range of _{1.2~3, <v q> max /} <v q> min is preferably set to 1.5 or less. Although the same clock period and the same pulse division method are used in the same zone, the linear velocity range that can be covered by the same pulse division method is generally limited to 1.5 times.

In the same zone, ω _q , T _q , α _i , β _i , Pe, Pb, and Pw are constant, and the physical length of the shortest mark n _min T _q <v _q > _ave is 0.5 μm. And T _q <v _q > _ave
Is substantially constant for all q such that 1 ≦ q ≦ p, and
m = n−1 or m = n−2,
α ₁ = 0.3 to 1.5,
α ₁ ≧ α _i = 0.2 to 0.8 (2 ≦ i ≦ m),
α _i + β _i-1 = 1.0 (3 ≦ i ≦ m),
0 ≦ Pb ≦ 1.5 (mW),
0.4 ≦ Pe / Pw ≦ 0.6.
Here, when m = n−1, α ₁ = 0.3 to 1.5, α _i = 0.2 to 0.5, and when m = n−2, α ₁ = 0.5 to 1. 0.5, α _i = 0.4 to 0.8 is preferable.

It is important to change the pulse division method according to the following rules. Pb, Pw, Pe / Pw ratio, α _i , β ₁ , β _m are variable for each zone, and at least α _i (i is at least one of 2 ≦ i ≦ m) from the outer peripheral zone toward the inner peripheral zone. Monotonically).
The change of α _i for each zone is preferably in increments of 0.1T or 0.01T.
Here, by adding a high frequency base clock generation circuit having a period of about 1/100 with respect to the reference clock period T _p in the outermost peripheral zone, T _q and the division pulse length in all the zones can be set to this base clock. It can be generated as a multiple.
In DVD, the reference clock frequency at 1 × speed is about 26 MHz, so a base clock frequency of up to about 2.6 GHz, usually a base clock frequency of about 260 MHz, which is an order of magnitude lower, is sufficient.

Further, when the maximum value of Pw in the recording area is Pw _max and the minimum value is Pw _min , Pw _max / Pw _min ≦ 1.2, Pe = Pw = 0.4 to 0.6, 0 ≦ Pb ≦ It can be set to 1.5 (mW). According to this, since the three types of power setting ranges can be limited, the power generation circuit can be simplified.
In the present invention, it is also possible to deal with all the linear velocities by changing only the pulse division method while keeping Pw, Pe / Pw ratio, and Pb constant. Also, β _m can be made constant regardless of the zone, and only α ₁ and α _m can be set as zone-dependent parameters. This is extremely useful for simplifying the recording pulse control circuit of the drive.

In the present invention, recording may be performed by virtually setting a zone on the recording medium from the radial position information of the optical head at the time of recording, or in accordance with address information and zone information previously described on the disk. A zone structure may be physically provided above. Whether it is virtual or physical, a recording pulse dividing method corresponding to the linear velocity determined by the zone may be selected.
Next, another example in which the optical recording method of the present invention is applied to the ZCAV system will be described.
The recording area is divided into p zones according to the radius, the innermost circumferential side is the first zone, the outermost circumferential side is the pth zone, and the angular velocity in the qth zone (where q is an integer of 1 ≦ q ≦ p) ω _q ,
The average linear velocity is <v _q > _ave , the maximum linear velocity is <v _q > _max , the minimum linear velocity is <v _q > _min , the reference clock cycle is T _q , and the time length of the shortest mark is n _min T _q To do.
In the ZCAV method, it is necessary to make the reference clock _Tq of the recording data smaller as the recording line density is shifted to the outer peripheral zone so that the recording linear density is substantially constant.
That is, T _q <v _q > _ave is substantially constant for all q _satisfying 1 ≦ q ≦ p.
_Tq is changed according to the zone. Here, “substantially constant” includes an error of about ± 1%.
In order to keep the maximum and minimum linear speeds in the same zone within a certain range,

Determine the width of the zone to satisfy. That is, (<v _q > _max− <v _q > _min ) is less than 10% of (<v _q > _max + <v _q > _min ), and the width of the q-th zone has an average radius <r _q
> A radius position of less than ± 10% of _ave is allowed.
More preferably, (<v _q > _max− <v _q > _min ) is less than 5% of (<v _q > _max + <v _q > _min ).
As for the zone width, the recording area may be equally divided for each radius, but may not be equally divided as long as this condition is satisfied. Although depending on the recording area width, the recording area having a width of 30 to 40 mm is roughly divided into 10 or more.

According to the study by the present inventors, even if the shortest mark length is about 0.4 μm, the jitter value is at a practical level if the expression (2) is satisfied.
The above two conditions are conditions for making the recording linear density constant, and hence the physical length of the mark or the channel bit length constant. The channel bit length is the length per channel bit along the track.
In order to more surely obtain playback compatibility with DVD, when the reference playback speed v is about 3.5 m / s and the reference clock period T is about 38.2 nsec, the fluctuation of the channel bit length vT is approximately ± It is preferable to be less than 1%.
In order to satisfy this condition in the ZCAV medium, the following equation (3)

Must be met. That is, (<v _q > _max− <v _q > _min ) is (<v _q > _max + <v _q
> _Min ), and the width of the q-th zone is allowed up to a radius position less than ± 1% of the average radius <r _q > _ave . For this reason, the recording area is divided into 200 or more zones. And,

And T _q <v _q > _ave is made substantially constant for all q _satisfying 1 ≦ q ≦ p. Here, “substantially constant” includes an error of about ± 1%.
As a result, pseudo-density recording can be performed regardless of the radius in spite of the ZCAV system, so that playback is possible even with the CLV system, and compatibility with a CLV DVD player is enhanced.
If necessary, the zone width may be narrower.

Now, an optical recording method for obtaining a recording density equivalent to that of a DVD under the above conditions will be described.
When recording and reproducing data by collecting light having a wavelength of 600 to 680 nm through an objective lens having a numerical aperture NA of 0.55 to 0.65 and condensing it on a recording layer through a substrate,
The innermost circumference of the recording area is in the range of radius 20 to 25 mm, the outermost circumference is in the range of radius 55 to 60 mm, the average linear velocity of the innermost circumference zone is 3 to 4 m / s,
The angular velocity in the q-th zone (where q is an integer of 1 ≦ q ≦ p) is ω _q and the average linear velocity is <
v _q > _ave , maximum linear velocity <v _q > _max , minimum linear velocity <v _q > _min , reference clock cycle T _q
If the time length of the shortest mark is n _min T _q ,
n is an integer of 1 to 14,
m = n−1,
ω _q , Pb and Pe / Pw are constant regardless of the zone,
T _q <v _q > _ave is substantially constant for all q _satisfying 1 ≦ q ≦ p, and

(I) In the first zone,
α ¹ ₁ = 0.3 to 0.8,
α ¹ ₁ ≧ α ¹ _i = 0.2 to 0.4 and constant regardless of i (2 ≦ i ≦ m),
α ¹ ₂ + β ¹ ₁ ≧ 1.0,
α ¹ _i + β ¹ _i−1 = 1.0 (3 ≦ i ≦ m), (ii) In the p-th zone,
α ^p ₁ = 0.3 to 0.8,
α ^p ₁ ≧ α ^p _i = 0.3 to 0.5 and constant regardless of i (2 ≦ i ≦ m),
When α ^p _i + β ^p _i-1 = 1.0 (2 ≦ i ≦ m), (iii) In other zones, α ¹ _i ≦ α ^q _i ≦ α ^p _i (2 ≦ i ≦ m) And α ^q ₁ is recorded as a value between α ¹ ₁ and α ^p ₁ .

When the innermost circumference of the recording area is in the range of radius 20 to 25 mm and the outermost circumference is in the range of radius 55 to 60 mm, the radial width of the recording area is about 30 to 40 mm. Then, the disk is rotated at a constant angular velocity so that <v ₁ > _ave = 3 to 4 m / s in the first innermost zone.
Recording is performed under the above conditions for the first zone and the pth zone, and α ¹ _i ≦ α ^q _i ≦ α ^p _i (2 ≦ i ≦) for the other zones (the q zone where 2 ≦ q ≦ p−1). m), and α ^q ₁ is a value between α ¹ ₁ and α ^p ₁ . In this case, the value of α ^q ₁ is desirably set in increments of 0.1T or 0.01T.
Preferably, α ¹ ₁ ≧ α ^q ₁ ≧ α ^p ₁ , where α ¹ ₁ > α ^p ₁ ).

Further, Pb, Pe / Pw, β ₁ and β _m are constant regardless of the zone, and if only α ₁ and α _i are changed depending on the zone, a wide linear velocity covering all the linear velocity of 3 to 8 m / s. Good overwrite characteristics can be obtained within a range.
Preferably, these Pe / Pw, Pb, Pw, β _m , (α ¹ ₁ , α ^p ₁ ), (α ¹ _c , α ^p _c )
Is previously described on the substrate by a prepit row or groove deformation, the drive may be able to select the optimum pulse division method and power for each recording medium and for each zone. These are normally recorded at a position adjacent to the innermost or outermost edge of the recording area. If the bias power Pb is the same as the reproduction power Pr, the bias power Pb may not be described. The groove deformation is specifically groove meandering (wobble).

Alternatively, the address information is recorded on the optical information recording medium in which the address information is previously recorded on the substrate by the prepit row or the groove deformation, together with the address information, an appropriate α _{1 at the} address.
And information about α _i may be included.
Accordingly, when accessing, the pulse division method information can be read out together with the address information, and the pulse division method can be switched, and the pulse division method suitable for the recording medium and the zone to which the address belongs without any special correction. Can be selected.

The above-described method of recording over the entire circumference of the disk while changing the recording pulse division method for each zone can also be applied to the ZCLV method (Zoneed CLV). A specific example will be described below.
The recording area is divided into a plurality of zones in the radial direction, and recording is performed at a constant linear velocity in each zone.
The ratio v _out / v _in the recording linear velocity v _out in the recording linear velocity v _in and the outermost zone in the innermost zone is 1.2 to 2,
α _i = 0.3 to 0.6 (2 ≦ i ≦ m) and β _m = 0 to 1.5,
Regardless of the linear velocity, m, α _i + β _i-1 (3 ≦ i ≦ m), α ₁ T, Pe / Pw, and Pb are constant, and α _i (2 ≦ i ≦ m) and Recording is performed by changing β _m .
In the ZCLV method, the recording area is divided into a plurality of zones in the radial direction in the same manner as the ZCAV method.

Therefore, when the recording method of the present invention is applied to the ZCLV system, when the linear velocities of the innermost zone and the outermost zone are V _in and V _out , respectively, the difference between V _in and V _out is reduced. By setting V _out / V _in to 1.2 to 2, the burden of linear velocity dependency on the medium is reduced.
Since the recording medium of the present invention can be recorded in a wide range of 3 to 8 m / s with only a slight change in the recording pulse dividing method, the ZCLV method of dividing into a relatively small number of zones can be applied.
At this time, in order to make the recording density equal regardless of the zone, the linear velocity V _q in each zone and the reference clock cycle T _q of the recording data in each zone are such that T _q <v _q > _ave does not depend on q. Almost constant.

In each zone, an optimized recording pulse division method is used. That is, α _i = 0.3 to 0.5 (2 ≦ i ≦ m) and β _m = 0 to 1.5, and m, α _i + β _i-1 (3 ≦ i ≦ m) regardless of the linear velocity. , Α ₁ T, Pe / Pw, and Pb are constant, and recording is performed by changing α _i and / or β _m according to the linear velocity.
In the CLV method, ZCAV method, or ZCLV method described above, the example in which the recording pulse division method is made variable according to the linear velocity at the time of overwriting mainly sets β _m constant regardless of the linear velocity, and generates a pulse. However, it is also possible to simplify the pulse generation circuit by positively changing β _m .

That is, when recording information with a plurality of recording mark lengths with a shortest mark length of 0.5 μm or less, with the crystal part in an unrecorded / erased state and the amorphous part in a recorded state,
Between the recording marks, irradiating recording light with an erasing power Pe that can crystallize amorphous,
When the time length of one recording mark is nT (T is a reference clock period, n is an integer of 2 or more),
The time length nT of the recording mark is

(Where m is the number of pulse divisions, m = n−k, and k is an integer satisfying 0 ≦ k ≦ 2.
Σ _i (α _i + β _i ) + η ₁ + η ₂ = n, η ₁ is a real number satisfying η ₁ ≧ 0, η ₂ is a real number satisfying η ₂ ≧ 0, and 0 ≦ η ₁ + η ₂ ≦ 2.0 .
α _i (1 ≦ i ≦ m) is a real number such that α _i > 0, and β _i (1 ≦ i ≦ m) is a real number such that β _i > 0.
α ₁ = 0.1 to 1.5, β ₁ = 0.5 to 1.0, β _m = 0 to 1.5, and α _i is 0.1 to 0.8 in i where 2 ≦ i ≦ m. And is constant regardless of i.
Note that α _i + β _i-1 is in the range of 0.5 to 1.5 in i where 3 ≦ i ≦ m, and is constant regardless of i. )
Divided in the order of
Within the time of α _i T (1 ≦ i ≦ m), the recording light with the recording power Pw of Pw> Pe for melting the recording layer is irradiated, and within the time of β _i T (1 ≦ i ≦ m). Is 0 <Pb ≦
Irradiate recording light with a bias power Pb of 0.2 Pe (however, in β _m T, 0 <Pb ≦ Pe),
Regardless of the linear velocity, m, α _i + β _i-1 (3 ≦ i ≦ m), α ₁ T, and α _i T (2 ≦ i ≦ m) are constant, and β _m becomes monotonous as the linear velocity decreases. This is an optical recording method that changes so as to increase.

First, in order to keep the recording density constant, the above-described ZCAV method or ZCLV method is applied, and the reference clock period T is changed in inverse proportion to the linear velocity.
Further, by making α _i + β _i-1 constant regardless of the linear velocity and i at least 3 ≦ i ≦ m, preferably 2 ≦ i ≦ m, the pulse generation circuit can be simplified, and α _i As the linear velocity decreases, the cooling rate of the recording layer can be increased. Usually, α _i + β _i-1 = 1.0.
In order to realize such a pulse division method, the width α ₁ T is synchronized with the reference clock period T (a constant delay may be added) in the explanatory diagram of the gate generation timing in FIG. One fixed-length pulse (Gate1) and a plurality of subsequent fixed-length pulses (Gate2) having a width α _i T (α _c T) are generated, while only Gate3 that determines the final off-pulse length β _m T is determined according to the linear velocity. Change it.

Here, when the maximum recording power at each recording linear velocity is Pw _max and the minimum recording power is Pw _min ,
Pw _max / Pw _min ≦ 1.2,
Pe / Pw = 0.4-0.6,
It is preferable that 0 ≦ Pb ≦ 1.5 (mW).
As described above, at least when the linear velocity at the time of overwriting is 5 m / s or less, in order to prevent thermal damage at the time of repeated overwriting, Σα _{i at} m = n−1.
It is preferable that Σα _i <0.5n when m = n−2.

Furthermore, the beta _m at the maximum line speed during overwriting beta ^H _m, the beta _m at the lowest linear velocity as beta ^L _m, between the beta _m in linear velocity during the overwriting beta ^L _m and beta ^H _m A recording method in which the Pb and Pe / Pw ratios are constant regardless of the recording linear velocity can be applied.
In this case, at least the values of Pe / Pw ratio, Pb, Pw, α ₁ T, α _i T, and (β ^L _m , β ^H _m ) are recorded in advance on the substrate of the medium by prepit rows or groove deformation. Therefore, the optimum pulse division method can be automatically selected, which is preferable.

Furthermore, if the maximum linear velocity is up to about twice the minimum linear velocity, an optical recording method in which β _m is constant regardless of the recording linear velocity while maintaining sufficiently practical signal quality is possible. is there.
There is a CLV system read-only DVD drive that performs rotation control by generating a data clock and a rotation synchronization signal based on a reference clock period obtained by reproducing a mark.

As described above, the medium on which the mark is recorded by the ZCAV method so that the shortest mark length or the channel bit length is substantially constant regardless of the recording radius is reproduced as it is by the reproduction-only DVD drive of this method. It is possible.
That is, it is possible to perform synchronous rotation control by a PLL (Phase Lock Loop) method so that the reference clock cycle T _q ′ of data generated from the recorded mark substantially coincides with the reference data clock Tr of the drive. Therefore, even if there is some fluctuation in the linear velocity or fluctuation in the channel bit length, it can be decoded as it is by the reproduction circuit.

In particular, EFM plus modulation data recorded so that the shortest mark length is almost constant at 0.4 μm in all zones achieves CLV rotation synchronization by PLL control from the rotation synchronization signal generated from the recorded mark. Is done. At the same time, a reference data clock Tr having a frequency in the range of 25 to 27 MHz is generated, and can be reproduced as a CLV recording medium based on this clock without being conscious of transition between zones.
Of course, if rotation synchronization is achieved so that the reference data clock becomes Tr / 2, reproduction at double speed is possible. For such a rotation synchronizing signal generation circuit by the PLL method, a method already known in a DVD player or DVD-ROM drive can be used as it is.

  The medium of the present invention can ensure reproduction compatibility with a DVD in all signal characteristics other than reflectance. For this purpose, in-groove recording is desirable and the push-pull signal of the groove is preferably small. This is because if the groove push-pull signal is large, the tracking servo signal in the DPD method used during reproduction is small. Therefore, it is necessary to make the groove depth shallower than λ / (8n) at which the push-pull signal becomes maximum. Here, λ is the reproduction light wavelength in the air, and n is the refractive index of the substrate. However, since a push-pull signal is normally used for tracking servo at the time of recording, it is not preferable that it is too small.

As for the reproduction signal characteristics, it is preferable that the modulation degree Mod is 0.5 or more in order to obtain a high CN ratio. However, Mod is (the amplitude of the envelope of the DC reproduction signal) / (the upper limit value of the envelope of the DC reproduction signal).
A preferable groove depth is d = λ / (20n) to λ / (10n). If it is too shallow than λ / (20n), the push-pull signal at the time of recording becomes too small to apply the tracking servo, and if it is deeper than λ / (10n), the tracking servo at the time of reproduction is not stable. For example, when the recording / reproducing wavelength is about 630 to 670 nm and the numerical aperture NA of the objective lens is 0.6 to 0.65, the groove depth is desirably in the range of 25 to 40 nm.

Further, in order to ensure the same capacity as that of the DVD, the groove pitch is set to 0.6 to 0.8 μm. Further, when the groove pitch is 0.74 μm, compatibility with the DVD is easily obtained.
The groove width is desirably 0.25 to 0.5 μm. If it is narrower than 0.25 μm, the push-pull signal becomes too small. When the width is larger than 0.5 μm, the width between the grooves becomes narrow, and it is difficult for the resin to enter during the injection molding of the substrate, so that accurate transfer to the groove-shaped substrate becomes difficult.
The reflectance of the medium of the present invention decreases after recording. In such a medium, in order to lower the reflectance in the groove, that is, RGa <RLa where RGa is the average reflectance in the groove after recording, and RLa is the average reflectance between the grooves after recording. Therefore, it is desirable that the groove width is narrower than the inter-groove width.

For example, for compatibility with DVD, when the groove pitch is 0.74 μm, the groove width is preferably narrower than 0.37 μm, which is a half of the groove width.
On the other hand, when the average reflectance in the groove before recording is RGb and the average reflectance between grooves before recording is RLb, RGb> RLb may be satisfied as long as RGa <RLa is satisfied. By setting the width to 0.4 to 0.5 μm, the width of the amorphous mark recorded in the groove can be widened to increase the degree of modulation and reduce the jitter in some cases.
These grooves may be provided with periodic deformations in order to access a specific unrecorded track and to obtain a synchronization signal for rotating the substrate at a constant linear velocity. In general, wobbles meandering in the cross-track direction are often formed. That is, if the groove meanders at a constant frequency fwo, a signal for rotation synchronization can be extracted by the PLL method by detecting the frequency.
The amplitude of the groove meander is preferably 40 to 80 nm (peak-to-peak value). If it is less than 40 nm, the amplitude is too small and the S / N ratio deteriorates. If it exceeds 80 nm, the upper and lower ends of the envelope of the recording signal shown in FIG. 6 contain a lot of low frequency components derived from the wobble signal, and the distortion of the reproduction signal is large. turn into.
When the wobble frequency is close to the recording data band, the amplitude is preferably 80 nm or less.

Furthermore, if the meandering frequency f _wo is used as a carrier wave and a frequency-modulated or phase-modulated meander is formed according to specific address information, the address information can be obtained by reproducing the meander.
If the groove meander is formed with the meander frequency f _wo constant, the reference clock signal T for data can be generated from the reference period T _w of the groove meander signal generated from f _wo , a multiple or a divisor thereof.
Normally, the wobble period is set to a frequency sufficiently lower than the frequency component of data to prevent mixing with the data signal component and to be easily separated by a band filter or the like. In particular, the fact that f _wo is lower by about 1 to 2 digits than the reference clock period of data has been put into practical use even for recordable CDs and the like.
In a medium used for the CLV method, after PLL rotation synchronization is achieved, f _wo is multiplied by about 1 to 2 digits to generate a data reference clock. Such data reference clock generated by methods generally, from the effects of the rotation synchronization fluctuation (0.1 to 1% of the f _wo), with the fluctuation of the same order as the data reference clock (frequency) Cheap. This exacerbates the window margin for data detection.

Therefore, in addition to the groove meandering signal, it is also effective to insert a prepit or a special wobble with a large amplitude for every fixed data length in order to correct the fluctuation of the data reference clock. On the other hand, if f _wo is the data reference clock frequency (1 / T) or a range from 1/100 to 100 times, after the rotation synchronization is achieved, the data reference clock is generated as it is based on the extracted wobble signal However, sufficient accuracy can be secured. That is,

And
In the previously described ZCAV method, the reference clock period T _q is preferably allowed to occur as a multiple or submultiple of the reference period Tw _q of the groove wobbling of each zone. That is, by changing the frequency f _wo for each zone and forming a meandering groove at a constant angular velocity, a reference clock generated as f _wo or its multiple frequency is generated as a data reference clock T _q. Can do.
At this time, if the wobble of the groove is set to a relatively high frequency that satisfies the equation (5), the data reference clock for each zone can be easily generated. For each zone, the reference clock T _q
The variable pulse dividing method can be generated in synchronization with this signal, and the position accuracy and fluctuation of each divided pulse can be reduced, which is preferable.
As an example of ZCAV system zone division, it is conceivable to make one round of the groove one zone. At this time, the groove has a wobble with a constant period regardless of the zone,
If the groove pitch is TP and the meandering cycle is T _wo , approximately

If this relationship is satisfied, a wobble with a constant period Tw ₀ is formed over the entire recording area, and a wobble increases every time the track goes around the circumference.
Then, Tw ₀ is that is an integer multiple of the reference clock period T, i.e. Tw ₀ = mT to (m is a natural number) has become, in the case of generating a reference clock from Tw _0, simply integer It is desirable that the reference clock generation circuit can be simplified because it is only a fraction. In this case, m may not be approximately a natural number, and a deviation of about ± 5% is acceptable.

That is, for TP = 0.74 μm, v ₀ = 3.5 m / s, T = 38.23 nse
When c and n = 1, m≈34.7. When approximately wobble period Tw ₀ = 35T, the number of wobbles included in one cycle increases by one.
In this case, although wobbles are introduced in the CLV system, the wobble phases of adjacent tracks are always aligned, so that there is an advantage that fluctuations in the reproduction amplitude of the wobble signal due to interference (beat) are small. is there.

The application examples of the present invention have been described above, but the present invention is effective in improving the linear velocity dependency and the recording power dependency in mark length recording of a phase change medium in general, and is limited to a rewritable DVD. is not.
For example, the present invention medium and recording method are also used when performing mark length modulation recording with a shortest mark length of 0.3 μm or less using a blue laser beam having a wavelength of 350 to 500 nm and an optical system with NA = 0.6 or more. It is valid. The shortest mark length is preferably about 10 nm or more in consideration of the stability of the mark.
In that case, it is necessary to pay attention to flattening the temperature distribution in the track crossing direction, and it is effective to make the thickness of the second protective layer as extremely thin as 5 to 15 nm.
When laser light having a wavelength of 350 to 450 nm is used, it is more preferably 10 nm or less.

  Further, the medium of the present invention may be applied to so-called land & groove recording in which recording is performed using both grooves as a track. Although there is a difficulty in satisfying the same recording characteristics between the land and the groove, the track pitch can be easily narrowed while the groove width is wide, which is suitable for high-density recording. By setting both the groove width GW and the inter-groove width LW to 0.2 to 0.4 μm, stable tracking servo performance can be obtained while being high density. Moreover, if the GW / LW ratio is 0.8 or more and 1.2 or less, the signal quality in both the grooves and between the grooves can be kept equal. In order to reduce the crosstalk, it is desirable to set the groove depth d = λ / (7n) to λ / (5n) or λ / (3.5n) to λ / (2.5n).

Examples will be shown below, but the present invention is not limited to the following examples as long as the gist thereof is not exceeded.
In the following examples, the substrate was produced by injection molding. The substrate was an injection-molded polycarbonate resin substrate having a thickness of 0.6 mm, and a substrate having a groove pitch of 0.74 μm, a width of 0.34 μm, and a depth of 30 nm formed on a spiral unless otherwise specified.
Unless otherwise specified, the groove had a wobble with a frequency of 140 kHz at a linear velocity of 3.5 m / s, and the amplitude of the wobble was about 60 nm (peak-to-peak value).

The groove shape was measured using an optical diffraction method approximating the U groove. The groove shape may be actually measured with a scanning electron microscope or a scanning probe microscope. In this case, the groove width at a position half the groove depth is used.
Unless otherwise specified, after a four-layer structure as shown in FIG. 5 (a) is formed on the substrate, a protective layer made of an ultraviolet curable resin is provided on the substrate by spin coating, and another layer having the same layer structure is formed. It was bonded to a 0.6 mm thick substrate. In the following examples and comparative examples, the first protective layer in FIG. 5A is referred to as a lower protective layer, and the second protective layer is referred to as an upper protective layer.

The recording layer immediately after film formation is amorphous, and the linear velocity is within a range from 3.0 to 6.0 m / s by a laser beam having a wavelength of 810 to 830 nm focused on a major axis of about 90 μm and a minor axis of about 1.3 μm. An appropriate linear velocity was selected, and light with an initialization power of 500 to 700 mW was irradiated to melt the entire surface and recrystallize to obtain an initial (unrecorded) state.
The composition of each layer was confirmed by combining fluorescent X-ray analysis, atomic absorption analysis, X-ray excitation photoelectron spectroscopy, and the like.
The film density of the recording layer and the protective layer was determined from the change in weight when the film was formed on the substrate as thick as several hundred nm. The film thickness was used after calibrating the fluorescent X-ray intensity with the film thickness measured with a stylus meter.
The area resistivity of the reflective layer was measured with a 4-probe resistance meter {Loresta FP, (trade name) manufactured by Mitsubishi Yuka (currently Dia Instruments)}.
Resistance measurement was performed on a reflective layer formed on an insulating glass or polycarbonate resin substrate, or on the reflective layer that is the uppermost layer after forming the four-layer structure (before UV curable resin protective coating) in FIG. .
Since the upper protective layer is a dielectric thin film and is an insulator, even if it has a four-layer structure, it does not affect the area resistivity measurement of the reflective layer. Further, the measurement was performed with the shape of a disk substrate having a diameter of 120 mm, which can be regarded as a substantially infinite area.
Based on the obtained resistance value R, the area resistivity ρs and the volume resistivity ρv were calculated by the following equations.

Here, t is a film thickness, F is a correction coefficient determined by the shape of the thin film region to be measured, and takes a value of 4.3 to 4.5. Here, it was set to 4.4.
Unless otherwise specified, a DDU1000 evaluation machine manufactured by Pulstec was used for recording / reproduction evaluation. The wavelength of the optical head is 637 nm, and the numerical aperture NA of the objective lens is 0.6 or 0.63. The beam diameter is about 0.90 μm and about 0.87 μm, respectively. The beam diameter corresponds to a region where the energy intensity of the Gaussian beam is 1 / e 2 or more of the peak intensity.

Recording was performed by the pulse division method shown in FIG. 10, where m = n−1 and α _i + β _i−1 = 1.0 (2 ≦ i ≦ m) unless otherwise specified. Pb was constant at 1.0 mW, the same as the reproduction power, at all linear velocities. Pe / Pw was constant at 0.5 unless otherwise specified. Pb was constant between 0.8 and 1.0 mW, and Pw was changed to measure the modulation factor and jitter.
The signal to be recorded was a random signal subjected to 8-16 modulation (EFM plus modulation) used in DVD, and the shortest mark length was 0.4 μm unless otherwise specified. Unless otherwise specified, the measurement was performed with only a single track recorded, so there was no effect of crosstalk.
Recording was performed at various linear velocities, such as 1 × speed and 2 × speed, with a DVD standard linear speed of 3.5 m / s as 1 × speed.

Reproduction was always performed at a linear velocity of 3.5 m / s, and jitter was measured after binarizing the reproduced signal after passing through the equalizer. Note that jitter refers to edge-to-clock jitter, and the measured value is expressed as a percentage of the reference clock period T. The characteristics of the equalizer are based on the DVD standard for playback. Reference clock period T = 38.2 nsec. It is preferable that a jitter of less than 10% (more preferably less than 8%) and a modulation degree of 50% or more, preferably a modulation degree of 60% or more are obtained with respect to (26.16 MHz). Furthermore, it is desirable that the increase in jitter after repeated overwriting is small and it is possible to maintain less than 13% with respect to T even after at least 100 times, preferably after 1000 times.
In addition, from the standpoint of ensuring compatibility with a read-only DVD, measurement with reproduction light at 650 to 660 nm is important. However, in the present invention, the wavelength only slightly affects the shape of the focused light beam. It has been confirmed that if the reproducing optical system is adjusted, the same jitter as the 637 nm optical system used in the present invention can be obtained even in the 660 nm optical system.

(Example 1 and Comparative Example 1)
As a recording layer, in order to compare the InGeSbTe system according to the present invention with a conventionally known InAgSbTe quaternary system, a medium having a recording layer composition and a layer configuration almost strictly excluding the composition of Ag and Ge is shown in Table 1. Prepared as follows.
The two recording layers have a composition within a range that can be considered to be sufficiently equivalent within a measurement error range except that Ag and Ge are replaced. The reason why the thickness of the lower protective layer is different is that the reflectance Rtop of the medium is adjusted to be the same. This correction is necessary because the refractive index of the recording layer is slightly different, but in order to compare the effect of thermal damage caused by the reproduction light with the same light absorption efficiency to the recording layer. Is a necessary correction. Since the recording layer thickness and the upper protective layer thickness are the same, the heat dissipation effect and thermal damage can be regarded as equivalent.
The substrate is a polycarbonate resin of 0.6 mm thickness, groove pitch 0.74 μm, groove width 0.34 μm, groove depth 27 nm, wobble frequency 140 kHz (linear velocity 3.5 m / s), wobble amplitude 60 nm (peak-to-peak). Value) is formed, and recording was performed in the groove.

When these two types of media were recorded with EFM plus modulation at a recording linear velocity of 3.5 m / s and T = 38.2 nanoseconds, good overwrite recording characteristics were exhibited. Overwrite recording conditions were not necessarily the conditions in which the characteristics of the respective disks were the best, but were performed under common conditions in which both characteristics were substantially equal as shown in Table-1.
That is, in the pulse division method shown in FIG. 10A, m = n−1, α _i + β _i−1 = 1.0 (2 ≦ i ≦ m), α _i = α _c = constant (2 ≦ i ≦ m), α ₁ = 0.5, α _c = 0.3, β _m = 0.5, Pw = 13.5 mW, Pe = 6.5 mW, and Pb = 0.8 mW.
The signal thus recorded was repeatedly irradiated with reproduction light, and the stability of the reproduction light was examined. After irradiation for a predetermined number of times with a predetermined reproduction light power Pr, the reproduction light power was sufficiently lowered to 0.5 mW, and jitter and the like were measured. The results are shown in FIG.
The medium of Example 1 did not show any deterioration due to the reproduction light up to 106 times at the reproduction light power of 1 mW. When the power is increased by 0.1 mW, the deterioration is gradually accelerated.

On the other hand, the medium of Comparative Example 1 gradually deteriorates after all the reproduction light having a reproduction light power of 1 mW or more suddenly increases jitter between the first 100 and 1000 times. Although the jitter value is high as a whole, initial jitter deterioration is fatal.
In Comparative Example 1, the degree of modulation was also reduced by the reproduction light, and it was settled down by about 10% after being irradiated about 100 times. Since the jitter increases rapidly in the initial stage, it is considered that the decrease in the modulation degree proceeds non-uniformly.
When the recorded media of Example 1 and Comparative Example 1 were left in an environment of 80 ° C./80% RH and subjected to an acceleration test, the characteristics of the disk of Example 1 changed almost completely after 250 hours. On the other hand, the recording signal of the disc of Comparative Example 1 almost completely disappeared. It can be seen that the amorphous mark is extremely unstable in the recording layer material having the composition of Comparative Example 1.

As described above, the disk of Example 1 is excellent in the stability of reproduction light resistance and stability over time as well as the initial overwrite recording characteristics. This indicates that the addition of an appropriate amount of Ge is very effective in an alloy system containing excess Sb in Sb _0.7 Te _0.3 .
The medium of Example 1 was subjected to an acceleration test in an environment of 80 ° C./80% RH. The acceleration test was conducted up to 2000 hours. The deterioration of the jitter of the signal recorded before the acceleration test was only about 1%.
Further, the modulation degree was 64% in the initial stage, but it hardly changed to 61% after the 2000 hour acceleration test. The reflectance was hardly changed at all. Jitter deterioration when recording was newly performed on an unrecorded part after 2000 hours was about 3%, but this is a level that does not cause any practical problem.
Further, in the medium of Example 1, the dependency of jitter on the recording pulse division method was examined in detail in the case of m = n−1 and m = n−2.

FIG. 13 is a contour diagram showing the α ₁ and α _c dependency of jitter when recording is performed at (a) m = n−1 and (b) m = n−2 at a linear velocity of 3.5 m / s, respectively. is there.
FIG. 14 shows contour lines showing the dependence of jitter on α ₁ and α _c when recording is performed at (a) m = n−1 and (b) m = n−2 at a linear velocity of 7.0 m / s, respectively. FIG. Pw, Pe, Pb, and β _m used for the measurement in each figure are shown above each figure.
At a linear velocity of 3.5 m / s, α ₁ = 0.
It can be seen that the lowest jitter (approximately 7% or less) is obtained in the vicinity of 7 to 0.8 and α _c = 0.35 to 0.40.
At a linear velocity of 7.0 m / s, α ₁ = 0.0 in both cases of m = n−1 and m = n−2.
It can be seen that the lowest jitter is obtained in the vicinity of 5 and in the vicinity of α _c = 0.40. For both α ₁ and α _c in the vicinity where the minimum jitter is obtained, the condition that Σα _i <0.5n is satisfied in both cases.
In this embodiment, a lower jitter value can be obtained by setting m = n−2 in both cases of linear speeds of 3.5 m / s and 7.0 m / s. Compared to the case of n−1, low jitter is obtained even for a large α ₁ .

Further, the dependence of jitter on the linear velocity of the medium of Example 1 was evaluated by changing the recording pulse division method as shown in Table 2 using an evaluation machine with NA = 0.63. The reference clock cycle T is inversely proportional to the linear velocity. In the pulse division method, m = n−1, α _i + β _i−1 = 1.0 (2 ≦ i ≦ m), and α _i = α _c = constant (2 ≦ i ≦ m). Pw, Pb, and Pe were constant regardless of the linear velocity. Here, in the pulse division method of Table 2, Σα _i <0.
5n is satisfied.
Good overwrite characteristics were obtained from the standard linear speed of DVD to about 2.5 times the normal speed. In this medium, the recording area is divided into 3 to 4 zones, and the recording pulse strategy is slightly changed for each zone, so that even in the CAV system, excellent overwrite characteristics are exhibited in the entire recording area.

  Similar results were obtained even when recording / reproduction was performed using an evaluator having a wavelength of 660 nm and NA = 0.65.

(Example 2)
On the substrate, a lower protective layer (ZnS) ₈₀ (SiO ₂ ) ₂₀ , a recording layer Ge _0.05 Sb _0.73 Te _0.22
The upper protective layer (ZnS) ₈₀ (SiO ₂ ) ₂₀ and the reflective layer Al _0.995 Ta _0.005 were provided by changing the thickness of each layer in various _ways . Table 3 shows the film thickness of each layer. All thin films were prepared by sputtering without releasing the vacuum.
The reflective layer was formed at an ultimate vacuum of 2 × 10 ⁻⁴ Pa or less, an Ar pressure of 0.54 Pa, and a film formation rate of 1.3 nm / second.
The volume resistivity was 55 nΩ · m, and the sheet resistivity was 0.28Ω / □.
Impurities such as oxygen and nitrogen are below the detection sensitivity in X-ray excitation photoelectron spectroscopy, and can be considered to be less than about 1 atomic% in total. The film density of the (ZnS) ₈₀ (SiO ₂ ) ₂₀ protective layer is 3.50 g.
/ Cm ³ , 94% of the theoretical bulk density of 3.72 g / cm ³ . The recording layer density was 90% of the bulk density. The thermal conductivity of the protective layer estimated from the thermal simulation was 3.5 × 10 ⁻⁴ pJ / (μm · K · nsec).

Recording (overwriting) was performed on the medium thus created using the pulse division method shown in FIG. 10A optimized for each layer structure of each medium at 1 × speed and 2 × speed. After that, the jitter after overwriting for the first time, 10 times, and 1000 times was measured. For the measurement, an optical system having a wavelength of 637 nm and NA = 0.63 was used for both recording and reproduction.
Table 3 summarizes the optimum pulse division method, jitter, Rtop, and modulation degree at 1 × speed for each medium.

In either case, mark length modulation recording with a minimum mark length of 0.4 μm can be performed at 1 × speed, and a large initial modulation degree is obtained.
When the thickness of the upper protective layer was 20 nm, both the initial jitter and the jitter after overwriting 1000 times were both less than 10%. When the thickness of the upper protective layer was 30 nm, the initial jitter was good, but the jitter increased slightly due to repeated overwriting, and the jitter became 10-12% after overwriting 1000 times. When the thickness of the upper protective layer was 40 nm, the initial jitter was 13% or more, and it was abruptly deteriorated by repeated overwriting and was 20% or more.
Furthermore, in Example 2 (h2) in which the recording layer thickness was increased to 30 nm, the initial recording jitter was 13% or more, and the deterioration of jitter due to repeated overwriting was remarkable.
In Example 2 (i2) in which the thickness of the lower protective layer was 45 nm, repeated overwrite durability was poor.
Further, a better jitter was obtained when the thickness of the reflective layer was 250 nm than when it was 200 nm. That is, in such a high-density mark length recording, it is understood that the “super quenching structure” is preferable.

Next, the recording power Pw dependence of jitter of the medium of Example 2 (g1) was evaluated. In the pulse division method, recording was performed at 1 × speed and 2 × speed with m = n−1, Pw = 14 mW, Pe / Pw = 0.5, and β _m = 0.5 in FIG. Thereafter, the dependence of jitter on α ₁ and α _c = α _i (2 ≦ i ≦ m) was evaluated.
At 2 × speed, α ₁ = 0.5, α _c = 0.4, β _m = β _n-1 = 0.5, Pw = 14 mW, and at 1 × speed, α ₁ = 0.7, α _c = 0.3 Β _m = β _n-1 = 0.5 and Pw = 14 mW. At this time, at double speed, Σα _i = 0.3n (n = 3), 0.33n (n = 4), 0.34n (
n = 5) and 0.38 n or less (n = 6 to 14). At 1x speed, Σα _i = 0.33n
(N = 3), 0.33n (n = 4), 0.32n (n = 5), and less than 0.32n (n = 6 to 14).
FIG. 15 shows the result. The recording power Pw dependence of the jitter after the first and 10th overwriting, and the recording power Pw dependence of the reflectance Rtop and the modulation degree Mod after the 10th overwriting are shown. (A) is a case of 2 × speed recording, and (b) is a case of 1 × speed recording. Note that Rtop corresponds to Itop in FIG. In the figure, DOW (Direct Overwrite) refers to overwriting.
Next, the overwrite durability was evaluated. FIG. 16 shows the result. For jitter, reflectance, and modulation degree, values up to 1000 times after overwriting are shown. (A) is a case of 2 × speed recording, and (b) is a case of 1 × speed recording. In either case, jitter gradually increased up to about 10 times, but stabilized after 10 times, and jitter, modulation degree, and reflectance were hardly deteriorated up to 1000 times.

Furthermore, this medium was overwritten with Pw = 14 mW by the same pulse division method as the above double speed (linear speed 7 m / s) except that the linear speed was 9 m / s and the reference clock period was 14.9 nsec. . A sufficient value of 30 dB or more was obtained for the erase ratio. The jitter was also good at less than 11%.
For the medium of Example 2 (g1), Pw = 14 mW, Pb = 1 mW, Pe / Pw = 0.5, β _m = 0.5, and constant α _{1 in} the linear velocity range of 3-8 m / s. and α alone altering _c good jitter was obtained. That is, α ₁ = 0 when the linear velocity is 3 to 5 m / s.
. 7, α _c = 0.35, linear speed 5-7 m / s, α ₁ = 0.65, α _c = 0.4, linear speed 7-8 m / s, α ₁ = 0.55, α _When changing to at least three stages such as _c = 0.45, a good jitter of less than 9% was obtained. If α ₁ and α _c are changed more finely in increments of 1 m / s, it is considered that better jitter can be obtained at each linear velocity.
Note that the best jitter was obtained when Pe / Pw was 0.4 to 0.5 at Pw = 11 to 14 mW. Further, when Pb exceeded 1.5 mW, the jitter deteriorated rapidly. Here, when Pb dependency was examined with Pe / Pw = 0.5, the best jitter was obtained when Pb was less than 1.0 mW. That is, Pb / Pe is preferably less than 0.2.

Next, Example 2 (g1) having an upper protective layer thickness of 20 nm and Example 2 (d2) having 40 nm are compared. For both media, the recording mark length dependency was measured at 1 × speed as follows.
Using an optical system with NA = 0.6, the dependency of jitter on the mark length was evaluated when the length of the 3T mark, which is the shortest mark in EFM plus modulation, was reduced from 0.5 μm. The recording linear velocity was constant at 3.5 m / s, the pulse division method was also constant as described above, and the mark length was changed by changing the reference clock cycle. However, when the shortest mark length is 0.46 μm or more, CLV control becomes difficult at a reproduction speed of 3.5 m / s due to restrictions on the apparatus, so the reproduction speed is set to 5 m / s. The shortest mark length of 0.4 μm corresponds to the read-only DVD standard.
FIG. 17 shows the result. (A) is the medium of Example 2 (g1), and (b) is the medium of Example 2 (d2).

It can be seen that the medium of Example 2 (g1) can be used with a jitter of less than 13% up to the shortest mark length of about 0.38 μm.
When an optical system with NA = 0.63 was used, it was possible to reduce jitter by about 2%. Further, when the equalizer at the time of reproduction is optimized, the jitter can be reduced by about 2%. In addition, if an optical system with NA = 0.65 is used, it is considered that sufficiently good jitter can be obtained even at 0.35 μm.
In the medium of Example 2 (d2), jitter having almost no problem was obtained when the mark length was 0.45 μm or more, but jitter suddenly increased when the mark length was less than 0.45 μm, and the jitter was 13% when the mark length was 0.40 μm. As a result, it became unusable.
Next, in order to evaluate the so-called tilt margin, after recording the EFM plus-modulated random pattern signal over a plurality of tracks on the medium of Example 2 (g1), the substrate is intentionally formed with respect to the optical axis of the reproduction laser beam. Tilt to measure the change in jitter during playback. The recording / reproducing optical system has NA = 0.6, the recording linear velocity is 1 × or 2 ×, and both are reproduced after 10 overwrites. FIG. 18 shows the measurement results. The tilt margin is ± 0.7 to 0.8 degrees in the radial direction and ± 0.5 to 0.6 degrees in the circumferential direction, which is a level with no problem in a normal drive.

<Acceleration test>
A random pattern EFM plus modulated was recorded on a part of the track of Example 2 (g1) with Pw = 13 mW, using the above-described optimal pulse division method, and jitter was measured. After that, the medium was subjected to an acceleration test under high temperature and high humidity of 80 ° C./80% RH. When the jitter of this track was measured again after 500 hours and 1000 hours after the acceleration test, it was only deteriorated by about 1% after 1000 hours.
Further, after 1000 hours of acceleration test, random patterns were recorded on other tracks under the same conditions as described above, and jitter was measured. As a result, deterioration of about 2% was observed. .
Further, recording was similarly performed at 1 × speed and 2 × speed, and the degree of modulation before and after the acceleration test for 1000 hours was evaluated under high temperature and high humidity of 80 ° C./80% RH. At 1 × speed, the initial modulation degree was 61%, and the modulation degree after the acceleration test was 58%. At double speed, the initial modulation degree was 60%, and the modulation degree after the acceleration test was 58%.
<Stability against reproduction light>
The medium of Example 2 (g1) was irradiated with the reproduction light with the power increased to 1.2 mW, but it did not deteriorate at all in about 10 minutes. Next, the power was set to 1.0 mW, and the reproduction light was repeatedly irradiated up to 1 million times, but the increase in jitter was less than 2%.

(Example 3)
Except for the recording layer composition being Ge _0.05 Sb _0.71 Te _0.24 , the same layer structure as in Example 2 was used.
Created media. Table 4 shows the film thickness and evaluation results of each layer. For the measurement, an optical system with NA = 0.63 was used.
Similar to Table 3, α ₁ , α _c , β _n-1 are optimized for each layer structure, and Pw, Pe
Also, jitter was evaluated by setting the jitter to be the lowest.
In either case, mark length modulation recording with a minimum mark length of 0.4 μm can be performed at 1 × speed, and a large initial modulation degree is obtained.
In Example 3 (a), as in Example 2 (a1), good characteristics were obtained when the recording linear velocity was 1 × and 2 ×, but at 9 m / s, 1 to 2 from Example 2 (a1). % Jitter was high.
In Examples 3 (a) to (f) in which the upper protective layer thickness was 30 nm, jitter of less than 10% was obtained, and after overwriting 100 times, it was less than 13%. In Examples 3 (g) to (i), where the upper protective layer thickness was as large as 40 nm, the jitter was only obtained to a value greater than 13%.

Example 4
The layer structure is a lower protective layer (ZnS) ₈₀ (SiO ₂ ) ₂₀ with a film thickness of 215 nm and a recording layer Ge _0.05.
Sb _0.69 Te _0.26 was 18 nm, the upper protective layer (ZnS) ₈₀ (SiO ₂ ) ₂₀ was 18 nm, and the reflective layer Al _0.995 Ta _0.005 was 200 nm. This recording layer composition provides good characteristics in recording at a linear velocity of 3 to 5 m / s, and is used for so-called 1 × speed. However, since the excess Sb amount is slightly less than those in Examples 2 and 3, it is excellent in stability over time, and is preferable for emphasizing storage stability of recorded information and deterioration due to repeated reproduction, that is, reproduction light durability. .
The following was evaluated with an optical system with NA = 0.6. The optimum pulse division method was determined as follows. At a recording linear velocity of 3.5 m / s, Pw = 13 mW, Pe / Pw = 0.5, and in FIG. 10, β _m = 0.5 is constant and α ₁ and α _c are changed to obtain the minimum jitter. The pulse division method was chosen. FIG. 19 shows the α ₁ and α _c dependence of jitter after 10 overwrites as a contour map of jitter. Almost the best jitter was obtained by setting α ₁ = 0.4 to 0.8 and α _c = 0.3 to 0.35. Based on this, α ₁ = 0.6, α _c = 0.35 was selected. At this time, Σα _i = 0.32n (n = 3), 0.33n (n = 4), 0.3n (n = 5), and less than 0.35n (n = 6 to 14).
The degree of modulation was 65%, a value comparable to that of a read-only DVD. Rtop is about 23%, but if it is practically 15% or more, it can be considered that reproduction is possible even with an existing reproduction-only drive.
Therefore, when image data was recorded on the recording medium of the present invention at Pw = 12.5 mW and a linear velocity of 3.5 m / s and reproduction was attempted with a commercially available DVD player, focus servo, tracking servo signal, and jitter were The same characteristics as a normal read-only DVD were obtained.

<Repeated overwrite durability>
FIG. 20 shows the dependency of jitter, Rtop, and modulation degree on the number of repeated overwrites at Pw = 12.5 mW. Even after overwriting 1000 times or more, it shows sufficiently stable characteristics.

<Acceleration test>
A random pattern EFM plus modulated was recorded on a part of the track of this medium with Pw = 13 mW and using the above-described optimal pulse division method, and jitter was measured. After that, the medium was subjected to an acceleration test under high temperature and high humidity of 80 ° C./80% RH. When the jitter of this track was measured again after 500 hours and 1000 hours after the acceleration test, it was only deteriorated by less than 0.5% after 1000 hours. The modulation degree was 65% in the initial stage and 63% after the acceleration test.
Further, after 1000 hours of the acceleration test, random patterns were recorded on other tracks under the same conditions as described above, and jitter was measured. As a result, deterioration of about 1% was observed. .

<Stability against reproduction light>
The medium was irradiated with reproduction light with the power increased to 1.3 mW, but it did not deteriorate at all in about 10 minutes. Next, the power was set to 1.0 mW, and the reproduction light was repeatedly irradiated up to 1 million times, but the increase in jitter was less than 1%.

(Example 5)
In the layer configuration of Example 2 (a1), the recording layer was Ge _0.05 Sb _0.75 Te _0.20 . Evaluation was performed with an optical system with NA = 0.6.
The best jitter was obtained at α ₁ = 0.4, α _c = 0.3, β _m = 0.5, Pw = 14 mW, and Pe / Pw = 0.5. The initial modulation degree was also sufficiently large. The jitter after overwriting 10 times was barely 10%, and remained below 13% after 1000 times.

<Acceleration test>
A random pattern EFM plus modulated was recorded on a part of the track of this medium with Pw = 14 mW and using the above-mentioned optimum pulse division method, and jitter was measured. After that, the medium was subjected to an acceleration test under high temperature and high humidity of 80 ° C./80% RH. When the jitter of this track was measured again after 500 hours of the acceleration test, it was only deteriorated by about 2%.
In addition, after 500 hours of the acceleration test, random patterns were recorded on other tracks under the same conditions as described above, and jitter was measured. As a result, deterioration of about 3% was observed. .

<Stability against reproduction light>
The medium was irradiated with reproduction light with the power increased to 1.0 mW, but it did not deteriorate at all in about 10 minutes. Next, the power was set to 1.0 mW, and the reproduction light was repeatedly irradiated up to 1,000,000 times. However, the increase in jitter was less than 3% and was maintained at less than 13%.

(Example 6)
In the layer configuration of Example 4, the recording layer was Ag _0.05 Ge _0.05 Sb _0.67 Te _0.23 . Evaluation was performed using an optical system with NA = 0.6.
Jitter pulse division method dependence (α ₁ and α _c ) measured at Pw = 13 mW, Pe / Pw = 0.5, m = n−1, β _m = 0.5 at a linear velocity of 3.5 m / s. As a result, FIG.
It became like the contour map shown in a). α ₁ = 0.6 and α _c = 0.35 were almost optimal. In this case, Σα _i = 0.32n (n = 3), 0.33n (n = 4), 0.33n (n = 5)
), Less than 0.35 n (n = 6 to 14).

FIG. 21 (b) shows the power dependence of jitter after the first, 10th and 1000th overwriting, and FIG. 21 (c) shows the power dependence of Rtop and modulation degree after the 10th overwriting. Good jitter was maintained in a wide recording power range until after overwriting 1000 times, and Rtop 18% and modulation degree 60% or more could be achieved.
FIG. 22 shows the change in jitter, Rtop, and modulation degree after overwriting 10,000 times at Pw = 13 mW. There was no deterioration at all except that the jitter increased by about 1% in the initial stage.
FIG. 23 shows the result of measuring the dependency of jitter on the shortest mark length by the same method as in Example 1. The shortest mark length was 0.38 μm, and the jitter was very good at less than 10%.
When the pulse division method with m = n−2 was also evaluated for this medium, α ₁ = 1.0, α _c = 0.5, β _m = 0.5, and FIG. Similar characteristics were obtained.
n = 3 in Σα _i = 0.48n, n = 4 in _{Σα i = 0.48n, Σα i =} 0.46n with n ≧ 5
It was -0.47n.

(Comparative Example 2)
In the layer configuration of Example 6, the recording layer was Ag _0.05 In _0.05 Sb _0.63 Te _0.27 .
When the dependence of jitter on the pulse division method was evaluated with Pw = 13 mW, Pe / Pw = 0.5, and β _m = 0.5 at a linear velocity of 3.5 m / s, the contour map shown in FIG. was gotten. α1 = 1.0, α _c = 0.5 is optimal, in this case, Shigumaarufa _i was constant at 0.5n irrespective of n.
The recording power dependency and the repetitive overwriting characteristics up to 1000 times are shown in FIGS. The jitter and power margin of the first recording were better than those of Example 5, but deteriorated by repeated overwriting, and became worse jitter after 1000 times.
Furthermore, when the reproducing light power was increased to 1 mW, the jitter deteriorated in about 5 minutes and increased to a few dozen percent. This difference cannot be explained by a recording sensitivity difference of 0.5 to 1 mW. The main cause of reproduction light deterioration is that the temperature rises to about 50 to 100 ° C., and it can be seen that the addition of Ge of the present invention is effective in improving the thermal stability of the amorphous mark.

(Comparative Example 3)
The layer structure is (ZnS) ₈₀ (SiO ₂ ) ₂₀ lower protective layer 90 nm thick, Ge ₂ Sb ₂ Te ₅ recording layer 21 nm, (ZnS) ₈₀ (SiO ₂ ) ₂₀ upper protective layer 23 nm, Al _0.995 Ta _{The 0.005} reflective layer was 200 nm.
In recording, the pulse division method shown in FIG. 10A was basically used, and fine adjustment was performed so as to obtain the best jitter at each mark length and linear velocity.
For this medium, as shown in FIG. 25, α ₁ = α _c = α ₀ = 0.3 to 0.4 is constant, and β _m = 1.0 is generally the best jitter. It was. Also, Pw = 13m
W, Pe / Pw = 0.4 (Pe = 5 mW), and Pb = 2.0 mW are optimum recording powers, and Pb / Pe = 0.4, which is higher. This is the recording of this comparative example. This is because the TL in FIG. 9 needs to be kept high to some extent in the layer.
Jitter was poor even when Pb was less than 1 mW, but jitter was also deteriorated when Pb was 3 mW or more.
Based on this pulse division method, further, precise pulse width adjustment of about 0.02 with respect to α ₀ was performed according to the mark length, and the mark length dependency was measured in the same manner as in Example 2. The results are shown in FIG. Moreover, the linear velocity dependency at the time of overwriting was measured. The results are shown in FIG.
The dependency on the linear velocity was such that the reference clock cycle was changed in accordance with the linear velocity, the shortest mark length was 0.4 μm, and reproduction was always performed at 3.5 m / s. As for the linear velocity dependency, the jitter after overwriting 10 times and the jitter when performing overwriting once after DC erasure are listed.
As shown in FIG. 26 (a), the shortest mark length is 0.4 μm and the jitter is 10%.
Further, as shown in FIG. 26B, the jitter deteriorates at a recording linear velocity of 5 m / s or more. However, the jitter is reduced by 2 to 3% or more in the recording after the DC erase once. From this, it is considered that the unevenness in temperature rise due to the difference in absorption rate between the so-called crystalline state and the amorphous state causes erasure failure or distortion of the shape of the amorphous mark, thereby deteriorating the jitter.

The jitter after overwriting at a linear velocity of 7 m / s was 20% or more, but it was about 15% in recording after DC erasure. Therefore, it is considered that the jitter at the high linear velocity is high because an appropriate pulse division method has not been selected.
This recording layer originally has high jitter due to coarse grains, but in addition to this, when the linear velocity is 5 m / s or more, the previous mark is not sufficiently erased at the time of overwriting, and jitter with recording after DC erasure The difference is clearly manifested.
The difference in jitter between the case of overwriting the medium of Example 2 (g1) at 7 m / s and the case of recording after DC erasure was less than 0.5%.
In the case of a recording medium using a GeTe-Sb ₂ Te ₃ pseudo binary alloy recording layer, such as Ge ₂ Sb ₂ Te ₅ , 5 to 6 m in a four-layer configuration including a protective layer / recording layer / protective layer / reflective layer. At a high linear velocity of / s or more, there is no problem with recording after DC erasure as described above, but jitter deteriorates during overwriting. For this reason, in order to reduce jitter, it is necessary to take measures such as correcting the absorptance by adding a light absorption layer or the like.

(Comparative Example 4)
In Example 2 (g1), the recording layer was made of Ge _0.15 Sb _0.64 Te _0.21 . Initial crystallization is very difficult. Initialization was finally performed by irradiating an initialization beam a plurality of times, and overwriting was performed to measure jitter. However, no matter how the pulse division method is changed within the range shown in FIG. % Or less jitter was not obtained. Moreover, when overwriting was repeated, the jitter increased by several percent between 10 times and 100 times.

(Comparative Example 5)
In the layer configuration of Example 2 (g1), the recording layer was Ge _0.05 Sb _0.80 Te _0.15 . At 7 m / s, α ₁ = 0.4, α _c = 0.3, β _m = 0.5, Pw = 14 mW, Pe / Pw =
The best jitter was obtained at 0.5, but the jitter was barely 11% after overwriting 10 times, and became 13% or more after 1000 times.

<Acceleration test>
A random pattern EFM plus modulated was recorded on a part of the track of this medium with Pw = 14 mW and using the above-mentioned optimum pulse division method, and jitter was measured. After that, the medium was subjected to an acceleration test under high temperature and high humidity of 80 ° C./80% RH. When the jitter of this track was measured again after 500 hours of the acceleration test, it deteriorated by about 3% and became 13% or more.
Further, after 500 hours of the acceleration test, random patterns were recorded on other tracks under the same conditions as described above, and jitter was measured. As a result, deterioration of about 5% was observed, and deterioration was rapid.

<Stability against reproduction light>
When this medium was irradiated with reproduction light with the power increased to 1.0 mW, the jitter increased by 3% after 10 minutes and was very unstable. In addition, the degree of modulation decreased and the marks tended to disappear.

(Example 7)
From the 1 × speed (linear velocity 3.5 m / s, reference clock period T = 38.2 nsec) to 2.25 × speed (7.9 m / s, T = 17 nsec) with respect to the medium of Example 2 (a1), α ₁ T = τ ₁ = 19 nsec, α _c T = τ _c = 11 nsec, all the linear velocities were constant, and only T was inversely proportional to the linear velocity, and an EFM plus signal was recorded. Further, to determine the _{α i + β i-1 =} 1.0 constant at beta _i. Only the final off-pulse period β _m was changed so as to become longer as the linear velocity was slower.
In such a pulse division method, τ ₁ = 19 in synchronization with the reference clock period T (a constant delay may be added) in the gate generation timing explanatory diagram of FIG.
One nsec fixed-length pulse (Gate 1) and τ _c = 11 nsec fixed-length pulse n−
Two (Gate2) may be generated, and only Gate3 that determines the final off-pulse length may be changed according to the linear velocity, which is preferable because the pulse generation circuit can be simplified. Further, in this embodiment, since the recording power Pw = 13.5 mW, Pe = 5 mW, and Pb = 0.5 mW are constant, the pulse generation circuit can be greatly simplified. Here, when the linear velocity is 5 m / s or less, Σα _i <0.47 n is satisfied, and thus thermal damage is sufficiently suppressed.
Table 5 summarizes the jitter values when β _m is changed at each linear velocity.
In the table, v represents a reference speed of 3.5 m / s. The wavelength of the pickup is 637 nm and NA = 0.63. The jitter value itself is slightly worse than when the pulse division method is made more flexible as in Example 2, but a value less than 10% is obtained from 1 × to 2.25 ×. It has been.
Here, if β ^H _m = 0.3 at double speed and β ^L _m = 0.6 (point surrounded by a square) at double speed, and β _m is changed in inverse proportion to the linear speed, then double speed It can be seen that a jitter of less than 10% can be obtained at each double speed. Furthermore, in this embodiment, although the margin of β _m is small, even if β _m = 0.2 is constant, jitter of less than 10% is obtained from 1 × speed to 2.25 × speed. In this way, it is possible to simplify the pulse generation circuit that can be varied according to the linear velocity.
In addition, if recording Pb, Pe / Pw, Pw, τ ₀ , τ _c , (β ^L _m , β ^H _m ) on the recording medium in advance by means of concave / convex pits or modulated groove meandering signals, the optimum recording conditions Can be automatically determined according to the linear velocity at the time of overwriting.

(Example 8)
The layer structure is a lower protective layer (ZnS) ₈₀ (SiO ₂ ) ₂₀ with a film thickness of 215 nm and a recording layer Ge _0.05.
Sb _0.69 Te _0.26 was 19 nm, the upper protective layer (ZnS) ₈₀ (SiO ₂ ) ₂₀ was 20 nm, and the reflective layer Al _0.995 Ta _0.005 was 200 nm.
With a linear velocity of 3.5 m / s, the pulse division method is α ₁ = 0.5, α _c = 0.35, β _m = 0.5,
Recording was performed by setting Pw = 11 mW, Pe = 6.0 mW, and Pb = 0.5 mW, and changing the reference clock period T to change the shortest mark length (3T mark length) from 0.4 μm to 0.25 μm. T = 38.2 nsec when the mark length of the 3T mark is 0.4 μm, and T = 19.1 nsec when the mark length is 0.2 μm. The recording laser wavelength is 637 nm and NA = 0.63.
Since this focused laser beam has a Gaussian distribution, it is possible to record at a higher density than the optical resolution by using only the high temperature portion at the center.
The recorded portion was reproduced with blue laser light having a wavelength of 432 nm, NA = 0.6, and power of 0.5 mW. This laser beam is generated by a nonlinear optical effect from a laser beam having a wavelength of about 860 nm. With this layer structure, a large modulation degree of 50% or more was obtained even at 432 nm.
Further, in FIG. 28, the jitter when reproducing with the optical system of 637 nm and NA = 0.63 used for recording and when reproducing with the optical system of 432 nm and NA = 0.6 is shown as the shortest mark length dependency. Indicated. In the measurement, the equalizer settings are optimized as much as possible at each measurement point. In this recording medium, it can be seen that good jitter of less than 13% is obtained even when the shortest mark length is 0.3 μm in blue laser light reproduction.

(Comparative Example 6)
In the layer configuration of Example 2 (a ₁ ), the recording layer was Ge _0.05 Sb _0.64 Te _0.31 .
Recording evaluation was performed with an optical system having a wavelength of 637 nm and NA = 0.63. At a linear velocity of 3.5 m / s, m = n−1, α ₁ = 0.4, α _c = 0.4, β _m = 0.4, Pb = 0.5 mW, Pe
= Overwrite recording was performed up to the 10th time by changing only Pw, with 4.5 mW being constant. FIG. 27A shows the recording power dependence of jitter at this time. In the figure, 1 write indicates the initial recording of an unrecorded disc, 1 DOW indicates the first overwrite, and 10 DOW indicates the 10th overwrite.
Next, with Pw = 8.5 mW, overwriting was performed up to the 10th time by changing only Pe. FIG. 27B shows the dependency of jitter on erasing power at this time.
In either case, good jitter was obtained in the first recording (1 write), but the jitter deteriorated rapidly after overwriting once. The recording layer composition in this comparative example is a composition richer than the straight line A in FIG. 3, and since the crystallization speed is slow, a sufficient erasure ratio cannot be obtained, and thus sufficient overwrite characteristics cannot be obtained. It is done.

(Example 9 and Comparative Example 7)
In the layer structure of Example 2 (a1), the recording layer composition was changed as shown in Table-6. The Ge amount is changed by co-sputtering a Ge _0.05 Sb _0.73 Te _0.22 target and Ge.
Using an optical system with a wavelength of 637 nm, NA = 0.63, m = n−1, Pb = 0.5 mW, β _m = 0.5, and α ₁ , α _c , Pw, Pe are changed and over 10 times We searched for conditions that would minimize jitter after writing.
The minimum jitter obtained with each recording layer composition was as shown in Table-6. Jitter increased as the amount of Ge added increased, and when Ge was 10 atomic% or more, the jitter at double speed was as high as 14%.
When this medium was subjected to an acceleration test under the conditions of 80 ° C. and 80% RH, Examples 9 (b) and (c) were slightly better than Example 9 (a). That is, when the signal recorded before the acceleration test was read out after 2000 hours of the acceleration test, the jitter was only deteriorated by about 1% in any of the examples 9 (a) to (c). .
In addition, the initial modulation degrees of Examples 9 (a) to 9 (c) were 61 to 63%, and 58 to 59% modulation degrees were obtained even after 2000 hours of acceleration test. The reflectance was hardly changed at all. In particular, in Examples 9 (b) and (c), the increase was within 0.5%.

Next, Ta was added by co-sputtering a Ge _0.05 Sb _0.73 Te _0.22 target and Ta. As a result, the best jitter was obtained when 1 to 2 atomic% of Ta was added to GeSbTe.

(Example 10 and Comparative Example 8)
In the layer configuration of Example 2 (g1), the recording layer was GeSbTe to which In was added. In is obtained by cosputtering InSbTe to a GeSbTe target. Each recording layer composition is Ge _0.05 Sb _0.74 Te _0.21 in Example 10 (a), In _0.023 Ge _0.048 Sb _0.719 Te _0.21 in Example 10 (b), and In _0.053 Ge _0.044 Sb _0.688 Te in Example 10 (c). _0.215, Comparative example 8 is _{in 0.118 Ge 0.041 Sb 0.617 Te 0.224} .
The results of evaluating the power dependency of the jitter of each medium are shown in FIGS. 29 (a), (b), (c), and (d). The upper row is for a recording linear velocity of 3.5 m / s, and the lower row is for 7.0 m / s.
All optical systems used were 637 nm, NA =. 63. When the linear velocity is 3.5 m / s, α ₁ = 0.6, α _c = 0.35, β _m = 0.5, and when 7.0 m / s, α ₁ = 0.4, α _c = 0.4 and β _m = 0.5. Pb was constant at 0.5 mW. Pe was fixed at two values, and only Pw was changed to measure the dependency of jitter on Pw. The Pw margin was greatly improved by adding about 2 to 5 atomic% of In. However, when it exceeds 10 atomic%, the jitter is worse than when it is not added.
Further, the jitter after 1000 overwrites was less than 10 atomic% in both linear speeds in Examples 10 (a) to (c), but in Comparative Example 8, both linear speeds were higher than 13%. .

<Acceleration test>
The medium of Example 10 (b) was subjected to an acceleration test in an environment of 80 ° C./80% RH. The acceleration test was conducted up to 2000 hours. The deterioration of the jitter of the signal recorded before the acceleration test was only about 1%.
The initial modulation degree was 61%, and a modulation degree of 57% was obtained even after 2000 hours of acceleration test. The reflectance was hardly changed at all.
Jitter deterioration when recording was newly performed on an unrecorded part after 2000 hours was about 3%, but this is a level that does not cause any practical problem.

(Example 11)
A disk having a recording layer of In _0.03 Ge0.05Sb _0.71 Te _0.21 in the layer structure of Example 2 (g1) was formed on a polycarbonate resin substrate having the groove shape shown in Table-7. In both cases, the groove pitch is 0.74 μm.

The wobble modulation method is binary phase modulation in which the carrier wave period Tw is 32 times the reference data clock period T = 38.2 nanoseconds. Here, the phase modulation wobble is to shift the phase of the wobble wave by π corresponding to 0 or 1 of the digital data signal as shown in FIG.
That is, an unmodulated carrier wave (cosine wave or sine wave) having a frequency f _c = 1 / T _w = 1 / (32T) is switched to a phase π by switching from 0 to 1 or 1 to 0 of the digital data for address. Just shift. Switching period T _d of the digital data 0 and 1 at low frequencies than T _w, since T _d is in a integral submultiple of T _w, be shifted phase [pi, wobble waveform varies continuously ing.
Unlike the frequency (FM) modulation used in ATIP (Absolute Time in Pregroove), this modulation method is preferable because the meandering frequency is constant and the period is modulated at a high frequency of 32T. The rotation synchronization of the disk is established with reference to the above, and the data clock can be directly generated in synchronization with the wobble clock.
In this way, in order to change the phase by modulation of digital data, for example, a ring modulator as shown in FIG. 31 is used. The digital data applies positive and negative voltages ± V corresponding to 0 and 1. At the time of producing the stamper master, exposure is performed while causing the laser beam for photoresist exposure to meander in the radial direction in accordance with a wobble waveform that is binary phase modulated between ± Vw voltages. At this time, the exposure beam can be meandered by applying the ring modulator output wave to the EO modulator.

Hereinafter, this will be described in some detail. In the figure, the period cos (2πf _c t) at the unmodulated carrier wave input terminal
When a signal V _w · cos (2πf _c t) is input, V _w · co
Two carrier signals of s (2πf _c t) and −V _w · cos (2πf _c t) appear. If the digital data input is positive (+ V), D ₁ and D ₁ ′ are conductive, and the carrier wave V _w · cos (2πf _c t
) Passes through D ₁ as it is, and the modulation appears at the output terminal. The carrier wave of −V _w · cos (2πf _c t) passes through D ₁ ′, and is then inverted by the transformer on the output side to become V _w · cos (2πf _c t), which is added to the output of D ₁ passing and V _w Obtain an output of cos (2πf _c t).
If the digital data input is negative (−V), that is, D ₂ and D ₂ ′ are conductive, the signal of V _w · cos (2πf _c t) is conducted to the lower side of the output transformer via the diode D _2. Therefore, the modulation is inverted at the output terminal to be −V _w cos (2πf _c t).

On the other hand, _{-V w · cos (2πf c t} ) and a carrier at the output of the input side transformer applied to the common mode input of the output transformer via a diode D ₂ ', as is polarity (-V _w · c
os (2πf _c t)) appears at the modulation wave output terminal. Accordingly, the carrier wave passing through the paths of the diodes D2 and D2 ′ is combined as −V _w · cos (2πf _c t), and the modulation appears at the output terminal.
In the case of a ring modulator, V _w · c is applied to the output terminal depending on whether the digital data input is positive or negative.
os (2πf _c t) or −V _w · cos (2πf _c t) is output.
The wobble waveform thus modulated is input to the EO modulator, and the exposure beam can be meandered.
In this embodiment, the wobble amplitude is all 60 nm (peak-to-peak value).
In the case of a medium for recording only in the groove, the preferred range of the groove depth is λ / (20n) = 20.20 for the recording / reproducing light wavelength λ = 637 nm and the refractive index n = 1.56 of the substrate. The upper limit is 5 nm and λ / (10n) = 40.8 nm.
For the evaluation of this medium, an optical system having a wavelength of 637 nm and NA = 0.63 was used.

As in the second embodiment, a recording pulse dividing method in which m = n−1, α _i + β _i−1 = 1.0 (2 ≦ i ≦ m), and α _i = α _c = constant (2 ≦ i ≦ m). When the linear velocity is 3.5 m / s, α _i = 0.
5, α _c = 0.3, β _m = 0.5, Pw = 13 mW, Pe = 6 mW, and at a linear velocity of 7 m / s, α ₁ = 0.4, α _c = 0.35, β _m = 0.5, Pw = 14 mW, and Pe = 7 mW.
First, recording was performed in the groove at a linear velocity of 3.5 m / s, and Rtop and the degree of modulation were measured. Further, the jitter of the recording signal was measured at 3.5 m / s and 7 m / s. The results are shown in Table-8.

First, Example 11 (k) had a very shallow groove with a depth of 18 nm, but the push-pull signal could hardly be detected and the tracking servo could not be applied. In addition, it is very difficult to form such shallow grooves uniformly in the production of a stamper. In practice, very large unevenness was observed in the tracking servo signal.
FIGS. 32 (a) and 32 (b) show the dependency of the modulation degree and Rtop on the groove shape. Examples 11 (h) to (j) have a groove with a depth of 42 nm, but the reflectance is significantly lower than that with a depth of 27 nm, which is not preferable because it is lower by 5% or more. The degree of modulation decreased particularly when the groove was thin. When the width was 0.23 μm, the degree of modulation was significantly reduced even at a depth of 35 nm.
In this example, the layer structure is the same. However, if the depth is 42 nm, the reduction in the degree of modulation becomes more conspicuous if the layer structure has a high reflectance in order to compensate for the decrease in the reflectance. That is, the groove having a depth of 42 nm is not suitable for recording in the groove.
When the groove depth is 40 nm or more, the wobble signal leaks significantly into the recording data signal when the groove width is less than 0.3 μm. Compared to when the groove width is 0.3 μm or more, the jitter deteriorates by 1 to 2% or more at a linear speed of 3.5 m / s, and deteriorates by 2 to 3% at a linear speed of 7 m / s.

(Example 12)
The layer structure is a lower protective layer (ZnS) ₈₀ (SiO ₂ ) ₂₀ with a film thickness of 65 nm and a recording layer Ge _0.05 S.
b _0.73 Te _0.22 was 16 nm, the upper protective layer (ZnS) ₈₀ (SiO ₂ ) ₂₀ was 20 nm, the first reflective layer Al _0.995 Ta _0.005 was 40 nm thick, and the second reflective layer Ag was 70 nm thick.
The lower protective layer to the first reflective layer are formed by sputtering without releasing the vacuum, and after the first reflective layer is formed, the atmosphere is released to the atmosphere and left for a few minutes, and then the second reflective layer is again formed by vacuum in the sputtering method. A film was formed.
After film formation of the second reflective layer, 4 μm of an ultraviolet curable resin was laminated as an overcoat layer by spin coating. The resulting discs were bonded together with the overcoat layer facing each other.
The first reflective layer was formed at an ultimate vacuum of 4 × 10 ⁻⁴ Pa or less and an Ar pressure of 0.55 Pa. The volume resistivity was 55 nΩ · m. Impurities such as oxygen and nitrogen were below the detection sensitivity in X-ray excitation photoelectron spectroscopy, and all were considered to be less than about 1 atomic%.
The second reflective layer was formed at an ultimate vacuum of 4 × 10 ⁻⁴ Pa or less and an Ar pressure of 0.35 Pa. The volume resistivity was 32 nΩ · m. Impurities such as oxygen and nitrogen were below the detection sensitivity in X-ray excitation photoelectron spectroscopy, and all were considered to be less than about 1 atomic%.
Using an optical system with a wavelength of 637 nm and NA of 0.60, a linear velocity of 3.5 m / s, α ₁ = 0.4
, Α _c = 0.35, β _m = 0.5, and the jitter after overwriting 10 times was measured. Pw = 11 mW, Pe = 6.0 mW, Pb = 0.5 mW A jitter of 6.5% was obtained.
When this medium was left to stand at a high temperature and high humidity of 80 ° C. and 80% RH for 500 hours and then recorded in the same manner, no deterioration was observed.

(Example 13)
A stamper with a groove pitch of 0.74 μm, a groove width of 0.3 μm, and a groove depth of 40 nm, in which a spiral groove having a wobble is formed, is used, and based on this, a polycarbonate resin having a diameter of 120 mm and a thickness of 0.6 mm The substrate was formed by injection molding.
As shown in Table 9, 36 mm with a radius of 22.5 mm to 58.5 mm was used as a recording area, and the recording area was divided into 255 bands (zones). Each band includes 191 tracks.
Since the band width is set so that the end of each band is exactly at the 191st track, each band width is not exactly 36/255. For this reason, the outermost end of the recording area is 58.54 mm.
The channel bit length is 0.133 μm, and a reference clock of 26.16 MHz (T = 38.23 nsec) is obtained at a linear speed of 3.49 m / s. The wobble period was set to be 9 times the channel bit length at the center radius of each band. Its physical period is 1.2 μm.
First, the total number of channel bit lengths and the total number of wobbles at the center radius of each band are calculated so that the number of channel bits or the number of wobbles included in one band is constant.

As shown in Table-9, the number of channel bits or the number of wobbles is constant with an accuracy of ± 1% at the beginning and end of the band. That is, recording with a constant linear density that is the same as that of the CLV method can be performed in the ZCAV method, and the reproduction-only DVD standard is sufficiently satisfied.
Based on the above assumptions, when the disc is rotated so that a linear velocity of 3.49 m / s is obtained at the center radius of each band, the wobble cycle is exactly nine times the reference clock cycle T = 38.23 nsec of DVD data. It becomes.
This medium is rotated so that the linear velocity is 3.49 m / s at the band center radius of the innermost peripheral band in Table 9, and used as a ZCAV system medium. The period of the carrier wave reproduced from the wobble of each band during CAV rotation is multiplied by 1/9 to generate the data reference clock T _q in each band, and the EFM plus modulated data is recorded based on the clock. .
When reproducing, if the rotation synchronization is achieved so that the data reference clock frequency generated from the recorded data is 26.16 MHz, the variation in channel bit length in each zone is less than ± 1%. Thus, the reproduction in the CLV mode can be performed without any trouble.

That is, the reference clock 26.16 MHz (T = 38.23 nsec) is generated by a crystal oscillator, the phase is compared with the data reference clock generated from the recorded data, and the two are synchronized. The rotation speed is finely adjusted by a normal PLL (Phase Locked Loop) control method.
Such rotation control by PLL control is currently performed by reproducing a DVD-ROM, and is useful in that the method can be applied as it is.

(Example 14)
In the layer configuration of Example 2 (a1), the reflective layer was Al _0.975 Ta _0.025 . The volume resistivity was 220 nΩ · m. A plurality of samples were prepared by changing the film thickness from 200 nm to 400 nm, and the jitter was measured using the optimum pulse division method in FIG. The best jitter of 12% was obtained at a film thickness of about 300 nm. Even if the reflective layer was made thicker or thinner than that, only worse jitter could be obtained.

(Example 15)
In the layer configuration of Example 11 (a), the thickness of the upper protective layer was 23 nm.
In-groove recording was performed on this medium. Using an optical system with a wavelength of 405 nm and NA = 0.65, an approximately circular beam with a spot diameter of about 0.5 μm (diameter at 1 / e ² intensity of a Gaussian beam) is generated and passed through a 0.6 mm thick substrate. Recorded and played back.
An EFM plus modulation signal in which the length of the shortest mark (3T mark) was 0.25 μm was recorded at a linear velocity of 4.86 m / s.
In the same recording pulse division method as in Example 2, m = n−1, α ₁ = 0.5, α _c = 0.38, β _m = 0.67, Pw = 9.5 mW, Pb = 0. 10 at 5 mW, Pe = 4.0 mW
When overwriting was performed once, the jitter was 10%.
It was found that recording / reproducing with a blue laser can perform higher quality recording than in the case of Example 7. Even a medium designed for the current red laser can be recorded and reproduced with the blue laser as it is to increase the density.

(Example 16)
A medium having a recording layer of Ga _0.05 Ge _0.05 Sb _0.68 Te _0.22 in the layer configuration of Example 2 (a1) was prepared. Initialization was performed in the same manner as in Example 2 (a1). For the measurement, an optical system having a wavelength of 637 nm and NA = 0.63 was used.
An EFM plus modulation signal in which the length of the shortest mark 3T was 0.4 μm was performed at a linear velocity of 3.5 m / s. In the same recording pulse strategy as in Example 2, m = n−1, α _i + β _i-1 = 1.0 (2 ≦ i ≦ m), α _i = α _c = constant (2 ≦ i ≦ m), Overwrite characteristics were evaluated by setting α ₁ = 0.5, α _c = 0.3, β _m = 0.5, Pw = 13.5 mW, Pe = 6.0 mW, and Pb = 0.5 mW. Initial recording (non-overwrite), 10-time overwriting, 100-time overwriting, 1000-time overwriting, jitter was good at 6.9%, 6.7%, 7.0%, and 7.3%, respectively. It was.
Further, similarly, α ₁ = 0.4, α _c = 0.35, β _m = 0.5 at a linear velocity of 7.0 m / s, Pw = 14.0 mW, Pe = 7.0 mW, Pb = 0. The overwrite characteristics were evaluated at 0.5 mW. Initial recording (non-overwrite), 10-time overwriting, 100-time overwriting, and 1000-time overwriting, jitter was good at 7.4%, 7.7%, 8.0%, and 8.5%, respectively. It was.
As for the degree of modulation, a value of 55 to 60% was obtained.
When this medium was left in an accelerated test environment of 80 ° C./80% RH for 1000 hours, recording was performed before the test. The deterioration of the jitter of the signal recorded before the acceleration test was less than 1%. Further, the modulation factor was 52 to 57%.

(Example 17)
As in Example 2, a wobble groove having a pitch of 0.74 μm is formed on a 0.6 mm thick polycarbonate resin substrate, and as shown in FIG. 5B, the reflective layer, the second protective layer, the recording layer, and the first protective layer are formed. The layers were formed in order.
The reflective layer Al _0.995 Ta _0.005 is 165 nm thick, the second protective layer (ZnS) ₈₀ SiO ₂ ) ₂₀ is 20 nm thick, the recording layer In _0.03 Ge _0.05 Sb _0.70 Te _0.22 is 16 nm thick, and the first protective layer (ZnS) ₈₀ (SiO ₂ ) ₂₀ was formed to a thickness of 68 nm by sputtering.
After that, a glass plate having a thickness of 0.6 mm was brought into close contact with the first protective layer. Initialization was performed by irradiating a laser beam of about 500 mW through a glass substrate at a linear velocity of 5 m / s.
Through this glass substrate, recording was performed by irradiating the recording layer with laser light using an optical system having a wavelength of 637 nm and NA = 0.6. Recording was performed on the far side of the unevenness as viewed from the laser incident side. This corresponds to the inside of the groove in Example 2.
An EFM plus modulation signal in which the length of the shortest mark 3T was 0.4 μm was performed at a linear velocity of 3.5 m / s. In the same recording pulse strategy as in Example 2, m = n−1, α _i + β _i-1 = 1.0 (2 ≦ i ≦ m), α _i = α _c = constant (2 ≦ i ≦ m), The overwrite characteristics were evaluated by setting α ₁ = 0.9, α _c = 0.35, β _m = 0.5, Pw = 12.0 mW, Pe = 6.0 mW, and Pb = 0.5 mW. After 10 overwrites, the jitter was 10.5% and the modulation factor was 61%.
Furthermore, similarly, α ₁ = 0.55, α _c = 0.40, β _m = 0.5 at a linear velocity of 7.0 m / s.
And Pw = 13.0 mW, Pe = 5.5 mW, and Pb = 0.5 mW, and the overwrite characteristics were evaluated. After overwriting 10 times, the jitter was 11.2% and the modulation factor was 61%.

  According to the stamper manufacturing method of the present invention, disk rotation synchronization is established with reference to a wobble clock, and a data clock can be directly generated in synchronization with the wobble clock.

The figure which shows the example of an amorphous mark shape. The figure which shows the reflectance change at the time of recording on the medium of an example of this invention. The GeSbTe ternary phase diagram showing the composition range of the recording layer of the medium of the present invention. The GeSbTe ternary phase diagram showing the range of the conventional GeSbTe composition. The schematic diagram which shows an example of the laminated constitution of the medium of this invention. The signal waveform figure for showing the relationship between signal strength, signal amplitude, and a modulation degree. The graph for demonstrating the 1st protective layer film thickness dependence of a reflectance. The figure which shows an example of the pulse division | segmentation method of a power ternary modulation recording system. The schematic diagram for demonstrating the time change of the temperature of a recording layer. The figure which shows an example of the pulse division | segmentation method of the power ternary modulation recording system suitable for mark length modulation recording. The conceptual diagram explaining the timing of 3 types of gate generation circuits for implement | achieving the pulse division | segmentation method of FIG. 6 is a graph showing the dependency of jitter on reproduction light power in Example 1 and Comparative Example 1. 6 is a graph showing the dependency of jitter on the recording pulse division method in Example 1. 6 is a graph showing the dependency of jitter on the recording pulse division method in Example 1. 6 is a graph showing the recording power dependence of jitter, reflectance, and modulation degree in Example 2. 6 is a graph showing the dependence of the jitter, reflectance, and modulation degree on the number of repeated overwrites in Example 2. The graph which shows the mark length dependence of the jitter in Example 2 (g1) and Example 2 (d2). 7 is a graph showing the dependency of jitter on the tilt angle of a substrate in Example 2. 10 is a graph showing the α1 and αc dependence of jitter after 10 overwrites in Example 4. 10 is a graph showing the dependency of jitter, Rtop, and modulation degree on the number of repeated overwrites in Example 4. (A) The graph which shows the pulse division method dependence of the jitter in Example 6, (b) The write power dependence of a jitter, and (c) The write power dependence of Rtop after 10 times overwriting and a modulation degree. 10 is a graph showing the dependency of jitter, Rtop, and modulation degree on the number of repeated overwrites in Example 6. 10 is a graph showing the mark length dependence of jitter in Example 6. (A) The graph which shows the pulse division method dependence of the jitter in the comparative example 2, (b) The write power dependence of a jitter, and (c) The write power dependence of Rtop after 10 times overwriting and a modulation degree. FIG. 10 is a diagram illustrating a pulse division method of a recording method used in Comparative Example 3. 10 is a graph showing the mark length dependency and linear velocity dependency of jitter in Comparative Example 3. The graph which shows the Pw and Pe dependence of the jitter in the comparative example 6. FIG. 10 is a graph showing the shortest mark length dependency of jitter in Example 8. The graph which shows the Pw dependence of the jitter in Example 10 and Comparative Example 8. FIG. The figure explaining the relationship between a digital data signal and a wobble waveform. The figure explaining the mechanism which modulates a wobble waveform with a digital data signal. 22 is a graph showing the dependency of the modulation degree and Rtop on the groove width in Example 11.

Claims (1)

A method of manufacturing a stamper in which a phase-modulated wobble signal is applied by meandering grooves,
A step of exposing the photoresist layer formed on the glass master by causing the laser light for photoresist exposure to meander in the radial direction in accordance with a wobble waveform that is binary phase modulated between ± Vw voltages;
In the step, the stamper manufacturing method is characterized by performing meandering of the laser beam for photoresist exposure by applying a ring modulator output wave to an EO modulator.

Images