-- substrate
...__ __
-+---_!11_111 --J. � recording layer reflective layer bleached area
Figure 5-2. Construction of a photochromic medium for multi-wavelength recording. The recording layer consists of a mixture of two kinds of photochromic compounds. Reflective layer is vacuum-deposited Ag film
The third and fourth components are due to linear and nonlinear crosstalk CH2 output using wavelength A-2 laser has a similar form.
These components can be expressed by reflectance
R/)1)
<;iS() =
_.!._ { R
( !J)R
(I) R ( 2) R c:�))
![)(' 4
�
1 + I + I + 1 '() . =
_.!._ { R
co) +R
Cl) -R
(2) -R
Ol)
IS 4
�
1 1 1 I '() =
_.!._ {R
(11) -R
(I) + R (2) -R U))
ILC' 4
�
I I I I '() 7,\'(' =
_!_
4{ � R
1 co) -R
I Cl) -R
1 (2) +R
I (3)\ }
(G-11)
(G-1 �)
(G-11)
h R(lt) > R<l) > Rc::) > R<') Were 1 1 1 I •
In general, the crosstalk is expressed as a ratio to the main signal. By introducing eq.
(5-9)
into R/fL\ we obtain the crosstalk intensity forCHI
as follow .Linear:
()1Lc
=�X=A.B 8x1C_\. (0) ·(I+
exp(-f3x1 �)XI-
exp(-J3\ 2 P2))
01s �X=A.B 8x/"' x (0) ·(I-
exp(-J3.n I�) XI+
exp(-jJ\ 2 P2))'
Nonlinear:
()INC
=�X =A.R 8.\'1
ex(0). (I-
exp(-J3.n �)XI-
exp(-J3 \ 2 p2))
20JS � X =A.R 8_\'1
ex (0). ( 1 -
exp(-J3 XI/�) XI +
exp(-J3 \ 2 p2)).
The expressions for
CH2
have similar forms.Linear:
()2LC'
=�X =A.B 6'_n
ex(0) . (1 +
exp(-f3):2 p2) XI -
exp(-J3.\ I/�)) 02s � x
=A.B 6'x2C'_\' (0) ·(I-
exp(-f3x:: P2) XI+
exp(-J3.n }�))'
Nonlinear:
O::.vc =
�X=A.B 8x2C'x (0) ·(I-
exp(-f3x2 P2) XI-
exp(-jJ_\.11�))
202S �X= . .J.B Ex2
e_,.(0). (I-
exp(-f3x:: p2) XI+
exp(-jJ_\ I PI) r
(G- 1 G)
(G- 1 (j)
(G- 17)
(G- 1H)
We consider the case in which the crosstalk in the reading process is neglected, in order to compare the theoretical expression with the experimental one using two diarylethenes with different absorption maxima. In this case, eqs.
( 5-I7)
and (5-18)
for CH2 are simplified as follows.Linear:
onlinear:
02Lc
_(1 +
exp(-j382 P2) XI -
exp(-j38J�))
()2S
-(1-
exp( -J3 B
2p2) XI+
exp( -jJ R I I�) r
02,vc (I
- exp(-J3 R
lPI))
20.,. 2 (I +
exp(-f381 �) )
'The main signal component in
CH2
output is as follows.Main signal:
54
(G- 1�))
(G-20)
Now we obtain the relationships between recording parameter
j382J>..,
and each component, as shown in Fig. 5-3. Under a condition off3ui)2=
1.2, which is about 3 dB below the saturation level of the main signal, we can estimate that the linear component of the crosstalk is about -2 dB and the nonlinear one is about -14 dB whenj382P2-
I.(a)
(b)
,--....""d o:l
"--"
'l] 1"'1
a
;-Q
::1bO 0
...
0 0l
(c)
,--....
""d o:l
"'-__./
c5
or.;-:z.
Q
bO 0 0 0l
0 - · · - - - - ... - .... . - Ju<o ... ... _ _ _ ....... ... .... ... ....... .... ...... ...
-20 L....o&.. ___ ..._ ___ _._ ___ _,
0
-10
-6 -10
-20
0
1 2 3
0
0
......
- .::-�.-- -
' - ' " ' " . . ."""'"""--·---�
1
1
fls2P2
2
1 -·
2
3
3
Figure 5-3. Relationships between the recording parameter
f382P2
and (a) the mam signal component, (b) linear crosstalk component, and (c) nonlinear crosstalk component inCH2
(a) Carrier level. We determined the typical recording conditions so that the carrier level was 3dB below the saturated value and
fJs2P2=1.2.
(b)
The ratio of the linear crosstalk component to the main signal component wa dependent onf381P1
andf382P2.
Under the conditions of/3R2P2=1.23
and/3H1F1=
I, the ratio was about-2
dB.(c) The ratio of nonlinear to linear crosstalk components was dependent on
f3 s2P2.
Thi value was always smaller than -6 dB by the factor of1/2
Under the condition of/Js2P2=1.23
and/381?1=1,
this ratio was about -14 dB.!5G
5.3 A Crosstalk Reduction Method
The output equation (5-1 0) indicates that the linear crosstalk component of one channel is included in the main signal component of another channel, and the nonlinear component is composed of the product of the main signals. The e cro talk components are reduced by an appropriate operation between the reading output of the multiplied channels. Now, we can express the reading output in the normalized form of
where
c;1
and (! are defined as the normalized amplitude of linear and nonlinear cro talk components corresponding to OtLriOts and OrNriOts, respectively.(h
ha a imilar form Here the DC component is ignored because that does not include any information The following are the crosstalk reduction operations.I
(}I = (}I -
61
. (}2'I
02 = (}:_
- 62
'01 .The linear crosstalk component can be eliminated by the choice of
The nonlinear crosstalk is reduced by the following operations.
II I I I
()I = (}I
- 0"1
. (}I (} 2 'II I I
02 = 02 -
0":_
'(}I (}:_(G-211)
(!}-2())
The nonlinear crosstalk component Wt + (1))_ can also be eliminated by the choice of o-1 and o-2, for example,
s1
-c;1s:.
(1- �1�2 y,
(f)-2H)In this case, however, other components, such as 2WJ + {1))_, WI+ 2WJ., are newly
R7
generated. Therefore, the parameters
a-1
anda-2
should be optimized so that the e newl_generated components become as small as possible. The intensities of the components in
CHI
output after operations(5-23), (5-24), (5-26)
and(5-27)
with the parameter choice(5-25)
and(5-28)
are expressed as follows.OJ
1 (main signal):OJ2 (linear):
2 OJ1 ±
OJ2
(nonlinear):OJ1
±2 OJ2
(nonlinear):(s� - ��s:. Xs:. - �2s1) 2 (1- �1�2)
(s� - ��s:. Xs2 - �:.s1) 4 (1- �I �2)
(s�- ��'2 Y 4 (1- �1 �2)
(G-2D)
Other components
(DC, 2 OJ1, 2 OJ2, 20J1
±2 OJ2
) are negligible because they are mailer than the above components. Because�
1 ands
1 are smaller than unity in general, s 12 ands
,2 are much smaller than unity. Expressions(5-29)-(5-32)
show that the cro talk components are reduced to a considerable extent by these operations.The crosstalk reduction method described above can be used for multi
wavelength recording even when the photochromic materials have broad absorption bands and poor wavelength selectivity. The method is also applicable to high-density multi-wavelength recording with the use of photochromic materials with narrower absorption bands.
5.4 Two-Wavelength Recording using Diarylethenes
In order to demonstrate the usefulness of the above cro stalk reduction method for multi-wavelength recording, the following experiments were carried ut. Two kind of diarylethenes, 2,3-bis(2-methylbenzo[b ]thiophen-3-yl) maleic anhydride (compound
A)
and 2-(1 ,2-dimethyl-3-indolyl)-3-(2,3,5-trimethyl-3-thienyl) maleic anhydride (compoundB),
were dispersed in poly( vinyl butyral) (PVB) ( 15 wt.% ) and u ed for the recording media. A Ag reflective layer was overcoated by vacuum evaporationThe photochromic behavior of the compounds is shown in Fig. 5-4. Compound A converts from the open to the closed-ring form upon irradiation with 405 nm light The photogenerated colored form converts to the open-ring form upon irradiation with light of wavelength longer than 500 nm. Bleaching of the longer-wavelength ab orption can also be induced by irradiation with the 515 nm Ar ion laser. On the other hand, compound
B
converts from the open to closed-ring form upon irradiation with 490nm light, and the closed-ring form returns to the open-ring form upon irradiation with light longer than 550nm. Bleaching can also be induced with the 633nm He-Ne laser.()_/�()
� LZs��sY
compound A
400 500 600
Wavelength (nm)
12
!! I
i i
I
400
\h.: I compound B
, ", .. ,,,,,I
12
� "
, ,/ I ',,
500 600
I \
Wavelength (nn1)
' ' ' '
700
Figure 5-4. Absorption spectra of 2,3-bis(2-methylbenzo[b ]thiophen-3-yl) maleic anhydride (compound
A)
and 2-(1,2-dimethyl-3-indolyl)-3-(2,3,5-trimethyl-3-thienyl) maleic anhydride (compoundB)
dispersed in poly( vinyl butyral) filmCompound A has a very weak absorption at 633nm. This indicates that the crosstalk in the writing/reading process using 633nm light is negligible. At 515nm, on the
!)9
other hand, both compounds A and B show absorption. Table 5-l summarizes the writing sensitivities Exr
rPr
r of compounds A and B dispersed in PVB. As can be seen from the table, the crosstalk induced in the writing process by 515nm light is as large as the main signal written by 63 3 nm light.Table 5-I. Recording sensitivities 6\·r
r/J.
rr of compounds A and B dispersed in poly( vinyl butyral).Compound
Wavelength (nm) 515 E.rr
tP
Xr (M-1cm-1) 1000Ar laser ND AOM
L...-1 5L...-1 _ 5nm ____.l-+l(
f1
HeNe laser NO AOM
A B
633 515 633
50 1300 1100
Photochromic Disc
PBS QWP DM1
OL
�
IHO+�,.LJ-HB-
....� :Y
DM2---ct-
PD1��6-33n_m
__
�I�[J
��----�QWP PBS
t
f2NO: neutral density filter
AOM: acous to-optical modulator f1 ,f2: recording signal
PBS: polarizing beam splitter
Q\NP: quarter-wave plate DM1 ,DM2: dichroic mirror OL: objective lens
LD: laser diode
APD1 ,APD2: avalanche photo diode
FE sensor: focal error sensor
FE sensor
controller
Figure 5-5. The apparatus for two-wavelength recording.
QWP PBS
780nm
LDFigure 5-5 shows the apparatus for the multi-wavelength recording experiment It is noted that a focal servo system using a near-infrared laser diode wa adopted in thi apparatus to avoid the destruction of signals during focusing. The 633nm and 515nm lasers were modulated to a certain frequency by acousto-optical modulator (AOM) and introduced to a pickup optical system through optical fibers which con erve polarization.
In the pickup system, these three laser beams were superposed using two dichroic mirrors (DM1, DM2) and focused into the same spot by a single objective len (OL). The diameter of this laser spot was about 2 Jiffi. In the reading proces , the beam reflected from the medium passed through the objective lens and were detected by PI photodetectors (PD). The crosstalk reduction operation was carried out between the e two PD outputs. The output signals before and after the crosstalk reduction operation were compared using a spectrum analyzer.
The writing and reading conditions are summarized in Table 5-Il The writing conditions correspond to
/3H2P2=j381P1=I.2.
Figure 5-6 shows the output power pectra of CHI and CH2. The crosstalk (100 kHz) in CHI is as small as about -24 dB The cro ·�talk in CH2 has linear (80 kHz) and nonlinear (20 kHz, 180 kHz) components who� e values are -2 dB and -14 dB, respectively.The circuit which simulates operations (5-23),(5-24),(5-26) and (5-27) were constructed using the differential amplifier (National Semiconductor Co., Ltd , LM6361) and analog circuit (Analog Devices Co., Ltd.� AD533). After the operation of eros talk reduction using this circuit, we obtained the power spectra shown in Fig 5-7. The crosstalk was considerably reduced to less than -25 dB by the operation.
Table 5-II. Write/read conditions
CHI CH2
Laser wavelength (nm) 515 633
Writing laser power (mW) 2.7 2.0
Reading laser power ( m W) 0.1 0.1
Recording signal frequency (kHz) 80 100
Relative speed (m/s) 1.4
CHI
RBW 3kHz VBW 100kHz 10dB/divCH2
RBW 3kHz VBW 100kHz 10dB/diV0 100
frequency (kHz)
200 0 100
frequency (kHz)
Figure 5-6. Output signal power spectra of CHI and CH2 before crosstalk reduction operations.
200
CHI
VBW 100kHz RBW 3kHz 1 OdB/divCH2
RBW 3kHz VBW 100kHz 1 Od8/d1v0 100
frequency (kHz)
200 0 100
frequency (kHz)
Figure 5-7 Output signal power spectra of CHI and CH2 after era talk reduction operations.
()2
200
5.5 Superlow-Power Readout of Multi-Wavelength Recording Medium We proposed the superlow-power readout method for a photon-mode optical memory in the Chap. 4. Since the SLP readout method simply detects the ab orption change of the medium at the readout laser wavelength, it can be applied easily to multi
wavelength recording. We have already analyzed the eros talk in multi-wavelength recording and proposed a crosstalk reduction method. In this section we pre� ent the results obtained by applying the SLP readout method to two-wavelength recording.
Two kinds of diarylethenes, 2,3-bis(2-methylbenzo[b ]thi phen-3-yl) maleic anhydride (compound A), and 2-(1-octyl-2-methyl-3-indolyl)-3-(2,3,5-trim thvl-1-thyenyl)maleic anhydride (compound B), as hown in Fig. S-8, were dispersed 111 polystyrene and used for the recording layer (thickness: 0. 7 J..lm). The Ag r flectiv Ia cr
v. as overcoated by vacuum evaporation and we obtained media having the stru ·ture shown in Fig. S-1. The photoreactions of compounds A and B has already been illustrated in Sec. 5 .4, and the write/read were carried out in the same manner
o-/\=o
{)_/ �s70sY �
compound A
400 500 600
Wavelength (nm)
400
' I I
'
\
\
()_/ -()
"\I
�s%{.<�}-1 �
()(;t I compound B
500 600
Wavelength
(n1n)Figure 5-8. Absorption spectrums of compound A and compound 8 dispersed in polystyrene films
700
Figure 5-9 shows the apparatus employed for the SLP two-wavelength recording.
Three lasers (515nm, 633nm and 780nm) were superposed using two dichroic mirror (DMs) and focused into the same spot by a single objective lens (OL). In the reading process, the readout beams reflected from the medium were divided by the DMs and
G:-3
detected by corresponding avalanche photodiode (APDs). Crosstalk reduction operations were carried out between these two APD outputs. We compared the output signals before and after the crosstalk reduction operations, using a spectrum analyzer.
The write/read conditions are summarized in Table 5-111 In order to clarify the influence of multi-wavelength crosstalk, we recorded the different frequencies 240 kHz in the Ar laser channel and 300 kHz in the HeNe laser channel. The readout la er powers were set at the superlow-power level.
optical fiber Ar laser ND AOM
l
L....-15 _ 15n _ m __. 1--fl.f(
f1 HeNe laser ND AOM
��63 _ 3nm___,l-+ �
f2
NO: neutral density filter
AOM: acousto-optical modulator f1 ,f2: recording signal
PBS: polarizing beam splitter QWP: quarter-wave plate DM1 ,DM2: dichroic mirror OL: objective lens
LD: laser diode
APD1 ,APD2: avalanche photo diode FE sensor: focal error sensor
FE sensor
controller
QWP PBS
780nm
LDFigure 5-9. The apparatus for SLP two-wavelength recording and readout.
Table 5-III. SLP 2-wavelength write/read condition
CHI Laser wavelength ( nm) 515 Writing power (mW) 2.0
Reading power (nW) 30
Recording signal frequency (kHz) 240
Relative speed (m/s) 1.4
CH2 633
I 7 20 300
Figures 5-l 0 and 5-11 show the output power spectra of the Ar Ia er and He e laser readout channels before and after the crosstalk reduction operations. In th Ar las r channel (CHI) the crosstalk was a considerably small value of -30 dB reflecting the lower absorption at 633 nm and a small absorption change at 515 nm for compound A Th CNR was an insufficient value of 40 dB (RBWlOk.Hz) owing to high Ia er noi e; an improved CNR value is expected for a low noise light source. On the other hand, the CNR and crosstalks after operations in the HeNe laser channel (CH2) were 50 dB and -40 dB, respectively. These results demonstrated that the SLP readout method allovv the coexistence of multiplexed recordings and crosstalk reduction operations.