A remarkably efficient azobenzene peptide for holographic information storage

Article abstract of DOI:10.1021/ja981402y

A new family of proline-based azobenzene peptides (DNO) for holographic information storage is reported. By use of polarization holography, it was found that gratings with extraordinarily high diffraction efficiency (up to 80%) can be recorded in hundreds

Full text of DOI:10.1021/ja981402y

4738
J. Am. Chem. Soc. 1999, 121, 4738-4743
A Remarkably Efficient Azobenzene Peptide for Holographic
Information Storage
Palle H. Rasmussen, P. S. Ramanujam, Søren Hvilsted, and Rolf H. Berg*
Contribution from the Risø National Laboratory, DK-4000 Roskilde, Denmark
ReceiVed April 24, 1998
Abstract: A new family of proline-based azobenzene peptides (DNO) for holographic information storage is
reported. By use of polarization holography, it was found that gratings with extraordinarily high diffraction
efficiency (up to 80%) can be recorded in hundreds of milliseconds in a ∼13-µm-thick film of dimer 10. This
represents a decrease of the response time by more than 2 orders of magnitude when compared to that of the
ornithine-based DNO dimer previously reported. Furthermore, it supports the expectation that increasing the
rigidity of the peptide backbone is crucial in the design of effective azobenzene peptides for optical recording.
Gratings recorded in 10 can be erased by circularly polarized light in a few seconds. It is also noted that,
unlike DNOs previously reported, 10 is soluble in common organic solvents and can be assembled by solution-
phase synthesis, which is mandatory for large-scale fabrication.
Introduction
Optical holography1,2 provides unique opportunities for
storing information at high density and opens up new applica-
tions that cannot be contemplated with other techniques. While
much attention will continue to be focused on the use of
inorganic crystals as erasable holographic storage media, organic
materials potentially offer major advantages, such as easy
processing and versatility in tailoring of properties. Desirable
organic materials should exhibit high diffraction efficiency and
fast response, high resolution and permanent storage until
erasure, easy erasure, and no high electric field to promote
recording or reading, and they should be amenable to large-
scale production. Azobenzene polymers,3-10 photorefractive
polymers,11-13 photopolymerizable liquid-crystalline mono-
mers,14,15 liquid crystals,16 and glasses17 have been investigated
Figure 1. Under illumination (hν ) 488 nm), the azobenzene
chromophore is reversibly isomerized through trans-cis-trans cycles.
As a consequence, the chromophore undergoes a number of random
reorientations. The optical transition moment axis (the arrow) lies
approximately parallel to the long axis of the trans-azobenzene.
* Corresponding author. E-mail: rolf.berg@risoe.dk.
(1) Gabor, D. Nature 1948, 161, 777-778.
to fulfill these criteria. The basis for the storage process is a
light-induced change in the material’s refractive index. In
azobenzene polymers, this change can be provided through
alignment of the chromophores. Thus, the azobenzene chro-
mophore reorients randomly through a number of laser-induced
trans-cis-trans photoisomerization cycles (see Figure 1) until
eventually its optical transition moment axis, which lies ap-
proximately parallel to the long axis of the trans-azobenzene,
lies in a plane perpendicular to the polarization direction of the
laser beam (see Figure 2). Oriented in this manner, the
azobenzene cannot absorb the light and isomerize, and, thus, it
will be in a stationary orientaion.
(2) Gabor, D. Science 1972, 177, 299-313.
(3) Eich, M.; Wendorff, J. H.; Reck, B.; Ringsdorf, H. Makromol. Chem.
Rapid Commun. 1987, 8, 59-63.
(4) Eich, M.; Wendorff, J. H. Makromol. Chem. Rapid Commun. 1987,
8, 467-471.
(5) Wiesner, U.; Antonietti, M.; Boeffel, C.; Spiess, H. W. Makromol.
Chem. 1990, 191, 2133-2149.
(6) Wiesner, U.; Reynolds, N.; Boeffel, C.; Spiess, H. W.; Stumpe, J.;
Mu¨ller, L.; Kreysig, D.; Hauck, G.; Koswig, H. D.; Ruhmann, R.; Ru¨bner,
J. Makromol. Chem. Rapid Commun. 1991, 12, 81-87.
(7) Natansohn, A.; Rochon, P.; Gosselin, J.; Xie, S. Macromolecules
1992, 25, 2268-2273.
(8) Xie, S.; Natansohn, A.; Rochon, P. Chem. Mater. 1993, 5, 403-
411.
(9) Hvilsted, S.; Andruzzi, F.; Ramanujam, P. S. Opt. Lett. 1992, 17,
1234-1236.
Recently, a novel class of oligopeptides with azobenzene
chromophores attached to an ornithine-based backbone (DNO,
Figure 3) was investigated as new material for holographic
information storage.18 The photochromic properties of azoben-
(10) Hvilsted, S.; Andruzzi, F.; Kulinna, C.; Siesler, H. W.; Ramanujam,
P. S. Macromolecules 1995, 28, 2172-2183.
(11) Ducharme, S.; Scott, J. C.; Twieg, R. J.; Moerner, W. E. Phys. ReV.
Lett. 1991, 66, 1846-1849.
(12) Liphardt, M.; Goonesekera, A.; Jones, B. E.; Ducharme, S.; Takacs,
J.; Zhang, L. Science 1994, 263, 367-369.
(16) Wiederrecht, G. P.; Yoon, B. A.; Wasielewski, M. R. Science 1995,
(13) Meerholz, K.; Volodin, B. L.; Sandalphon; Kippelen, B.; Peygham-
barian, N. Nature 1994, 371, 497-500.
270, 1794-1797.
(17) Lundquist, P. M.; Wortmann, R.; Geletneky; C.; Twieg, R. J.; Juricj,
(14) Zhang, J.; Sponsler, M. B. J. Am. Chem. Soc. 1992, 114, 1506-
1507.
M.; Lee, V. Y.; Moylan, C. R.; Burland, D. M. Science 1996, 274, 1182-
1185.
(15) Zhang, J.; Carlen, C. R.; Palmer, S.; Sponsler, M. J. Am. Chem.
Soc. 1994, 116, 7055-7063.
(18) Berg, R. H.; Hvilsted, S.; Ramanujam, P. S. Nature 1996, 383, 505-
508.
10.1021/ja981402y CCC: $18.00 © 1999 American Chemical Society
Published on Web 05/11/1999

Azobenzene Peptide for Holographic Information Storage
J. Am. Chem. Soc., Vol. 121, No. 20, 1999 4739
Figure 2. A stationary orientation is obtained when the optical
transition moment is oriented in a plane perpendicular to the polarization
direction (Eˆ) of the laser beam.
Figure 4. Conceptual model that illustrates the (speculative) rationale
behind the DNO design. Assuming that Eˆ is oriented perpendicular to
the plane of the paper, it is then clear that any axis (arrow) lying in the
plane of the paper, or in a plane parallel to the paper, will be in a
stationary orientation. Thus, in the three cases shown, i.e., the single
azobenzene with its optical transition momemt lying along the y-axis,
the two neighboring azobenzenes held together in a parallel orientation,
and the two neighboring azobenzenes stacked on top of each other in
a nonparallel orientation, the molecule in question is in a stationary
orientation. In all three cases, the molecules can be rotated in the plane
of the paper while retaining themselves in a stationary orientation.
Furthermore, the single azobenzene can be rotated in any position
around its axis while retaining itself in a stationary orientation. Likewise,
the two azobenzenes held together in a parallel orientation can be rotated
around either one of the axes while retaining the molecule in a stationary
orientation. However, if the molecule with the two azobenzenes oriented
in a nonparallel manner (i.e., the DNO molecule) is rotated in any
position, other than 180°, around either one of its axes, it leads to a
situation where the neighboring axis will no longer be oriented in a
stationary orientation because it would still be able to undergo
photoisomerization cycles. As a result, there should be considerably
fewer (“one dimension” less) stationary orientations possible in this
case. In other words, with the DNO molecule, a more uniform alignment
of the molecule might be expected than in cases where the azobenzenes
either are not stacked or are stacked parallel to one another.
diffraction efficiency, which indicates how much of the light
entering the sample is diffracted by the holographic gratings,
can be recorded in films only a few micrometers thick of a
simple dimer (n ) 1, Figure 3) or longer oligomers. No electric
field is required to obtain the high diffraction efficiency, and
the holograms, which have lifetimes on the order of years at
room temperature, can be erased by exposure to circularly
polarized light. In the case of the dimer, it takes about 5 min to
obtain a diffraction efficiency of 75%. From both scientific and
practical viewpoints, shortening the response time would be
highly desirable. In the present study, we sought to modify the
backbone in a manner that would permit faster recording.
Figure 3. Chemical structures of ornithine-based DNO, DNA, and
proline-based DNO.
zene-containing polypeptides, introduced by Goodman and
Kossoy more than three decades ago,19 have been investigated
in numerous studies,20 and, in regard to holographic storage,
spiropyran-containing polypeptides have been investigated as
well.21
In analogy to the design of peptide nucleic acids (PNA),22-24
DNO25 has a molecular geometry similar to that of DNA. The
basic idea underlying the design of DNO was to construct a
molecule in which neighboring azobenzene groups are posi-
tioned in such a way that the transition moment axes are
nonparallelly oriented in planes parallel to one another (see
Figure 4). This was expected to lead to a more uniform
alignment of the chromophores and the backbone since there
should be fewer stationary orientations possible. The results
showed that holographic gratings with extraordinarily high
Experimental Section
1
General Methods. H NMR and 13C NMR spectra were recorded
on a Bruker DPX-250 spectrometer at 300 K. Chemical shifts are
expressed as δ values in parts per million (ppm) relative to an internal
standard of TMS. NMR spectra were recorded in CDCl3 unless
otherwise indicated. Melting points and glass transitions were deter-
mined by differential scanning calorimetry (DSC) of the second heating
trace at a rate of 10 °C/min.
Materials. All reagents were obtained at highest commercial quality
and used without further purification, unless otherwise stated. All
solvents used were of superpurity grade.
N-Ac-L-cis-Hyp-OMe (2). To a stirred solution of HCl‚H-L-cis-Hyp-
OMe (1, 4.0 g, 22.0 mmol) in dioxane-H2O (1:1, 20 mL) at 0 °C was
added NaHCO3 (4.0 g, 47.6 mmol) in small portions over 10 min,
together with acetic anhydride (2.32 mL, 24.5 mmol) in dioxane-H2O
(1:1, 5.0 mL), dropwise over 30 min. After 2 h at 25 °C, the resulting
solution was concentrated in vacuo, diluted with saturated aqueous NaCl
(20 mL), and extracted with CHCl3 (10 × 15 mL). The pooled extracts
were dried (MgSO4) and concentrated to a clear oil. Yield: 4.0 g (97%).
1H NMR: δ 1.95 (s, 1H), 2.05 (s, 3H), 2.30 (m, 2H), 3.60 (m, 2H),
3.71 (s, 3H), 4.36 (m, 1H), 4.42 (m, 1H). 13C NMR: δ 21.95, 37.06,
52.47, 56.16, 57.22, 70.50, 169.87, 173.88.
(19) Goodman, M.; Kossoy, A. J. Am. Chem. Soc. 1966, 88, 5010-
5015.
(20) For a recent review, see: Pieroni, O.; Fissi, A.; Popova, G. Prog.
Polym. Sci. 1998, 23, 81-123.
(21) Cooper, T. M.; Tondigli, V.; Natarajan, L. V.; Shapiro, M.;
Obermeier, K.; Crane, R. L. Appl. Opt. 1993, 32, 674-677.
(22) Nielsen, P. E.; Egholm, M.; Berg, R. H.; Buchardt, O. Science 1991,
254, 1497-1500.
(23) Egholm, M.; Buchardt, O.; Nielsen, P. E.; Berg, R. H. J. Am. Chem.
Soc. 1992, 114, 1895-1897.
(24) Egholm, M.; Nielsen, P. E.; Buchardt, O.; Berg, R. H. J. Am. Chem.
Soc. 1992, 114, 9677-9678.
(25) We have chosen to call this type of structures DNO since the
azobenzene peptides used in the previos work (ref 18) were all based on
diamino acid NR-substituted oligopeptides.

4740 J. Am. Chem. Soc., Vol. 121, No. 20, 1999
Rasmussen et al.
Figure 5. Experimental setup for polarization holography. M1, M2,
and M3 are mirrors; HWP, half-wave plate; QWP, quater-wave plates;
PBS, polarization beam-splitter; D, detector.
Figure 6. UV-visible absorption spectrum of 10 coated as a very
thin film (approximately 100 nm thick) on a glass substrate.
phases were concentrated to an orange solid, which was dissolved in
chloroform-methanol (4:1, ∼50 mL) and purified on basic alumina
with chloroform (350 mL), 10% citric acid (350 mL), and methanol-
Chart 1
1
chloroform (1:4, 250 mL) as eluents. Yield: 6.0 g (94%). H NMR
(DMSO-d6): δ 2.01 (s, 3H), 2.30 (m, 1H), 2.60 (m, 1H), 3.80 (d, 1H,
J ) 11.52 Hz), 3.95 (d, 1H, J ) 4.13 Hz), 4.33 (t, 1H, J ) 7.96 Hz),
5.28 (bs, 1H), 7.22 (d, 2H, J ) 8.77 Hz), 7.96 (bd, 4H, J ≈ 8.6 Hz),
8.05 (d, 2H, J ) 8.41 Hz). 13C NMR (DMSO-d6): δ 22.09, 34.62,
52.78, 57.12, 76.29, 112.63, 116.15, 118.46, 122.91, 125.30, 133.73,
146.41, 154.10, 160.33, 168.54, 172.91.
N-Boc-L-trans-Pro(4-{4-[(E)-2-(4-cyanophenyl)-1-diazenyl]phe-
noxy})-OMe (7). To a stirred solution of methyl N-Boc-cis-4-hydroxy-
L-prolinate 6 (5.0 g, 20.38 mmol), compound 3 (5.92 g, 26.50 mmol,
1.5 equiv), and PPh3 (5.88 g, 22.40 mmol, 1.1 equiv) in dry THF (100
mL) was added at 0 °C diethyl azodicarboxylate (DEAD) (3.53 mL,
22.40 mmol, 1.1 equiv), and the mixture was stirred at 23 °C for 16 h
under argon. The mixture was concentrated in vacuo to a red oil, which
was dissolved in chloroform (∼15 mL) and run through basic alumina
using chloroform as eluent. The resulting solution was concentrated in
vacuo to an orange solid, which was purified by flash chromatography
on a silica column (3 × 18 cm) using methylene chloride-diethyl ether
1
(9:1) as eluent. Yield: 7.5 g (82%). H NMR: δ 1.40 (s, 9H), 2.55
4-[(E)-2-(4-Hydroxyphenyl)-1-diazenyl]benzonitrile (3). To a solu-
tion of 4-aminobenzonitrile (7.08 g, 0.06 mol) in 99% ethanol (60 mL)
was added 6 N HCl (100 mL), causing immediately a precipitate. The
ethanol was removed in vacuo at maximum 55 °C. After the mixture
was stirred for 10 min at 0 °C, a precooled (0 °C) solution of NaNO2
(4.14 g, 0.06 mol) in H2O (60 mL) was added dropwise, constantly
keeping the temperature below 5 °C. The cold mixture was added
dropwise to a solution of phenol (5.64 g, 0.06 mol) in 3 N NaOH (200
mL), constantly keeping the temperature between -5 and -1 °C. The
reaction mixture was stirred for 1 h and left for 16 h. The precipitate
was washed with H2O, dissolved in 50% aqueous ethanol (800 mL),
and then precipitated with 12 N HCl, filtrated, and air-dried. Yield:
11.3 g (84%), mp 190 °C. 13C NMR (DMSO-d6): δ 112.15, 116.10,
118.42, 122.64, 125.48, 133.41, 145.34, 154.31, 162.08.
(m, 2H), 3.75 (s, 3H), 3.85 (m, 2H), 4.55 (m, 1H), 5.02 (m, 1H), 6.94
(d, 2H, J ) 8.6 Hz), 7.80 (d, 2H, J ) 8.5 Hz), 7.95 (bd, 4H, J ) 8.5
Hz). 13C NMR: δ 28.02, 35.20, 51.94, 52.90, 57.58, 74.72, 80.12,
112.91, 115.53, 119.73, 122.76, 125.13, 132.92, 146.67, 154.14, 159.96,
170.64, 172.57.
N-Boc-L-Ala-L-trans-Pro(4-{4-[(E)-2-(4-cyanophenyl)-1-diazenyl]-
phenoxy})-OMe (8). Compound 7 (4.0 g, 8.88 mmol) was stirred in
trifluoroacetic acid-methylene chloride (1:1, 100 mL) at 0 °C for 1.5
h. The mixture was concentrated in vacuo and coevaporated with
benzene-methylene chloride (1:1, 3 × 50 mL) to a red oil, which was
dissolved in dry THF (100 mL). Boc-L-Ala-OH (1.68 g, 8.88 mmol),
1-hydroxybenzotriazole (HOBt) (1.20 g, 8.88 mmol), and Et3N (5.1
mL, 26.6 mmol, 3.3 equiv) were added, and the mixture was cooled to
0 °C. 1-Ethyl-3-(3′-dimethylaminopropyl)carbodiimide hydrochloride
(WSC‚HCl) (2.16 g, 11.28 mmol, 1.27 equiv) was added, and the
mixture was stirred for 16 h at 20 °C under argon. The mixture was
concentrated in vacuo to a red oil, which was dissolved in chloroform
(100 mL) and washed with water (3 × 100 mL). The organic phase
was concentrated to a foam, which was purified by flash chromatog-
raphy on a silica column (5 × 18 cm) using chloroform-methanol
N-Ac-L-trans-Pro(4-{4-[(E)-2-(4-cyanophenyl)-1-diazenyl]phe-
noxy})-OMe (4). To a stirred solution of compound 2 (3.90 g, 20.8
mmol), compound 3 (7.10 g, 31.7 mmol, 1.5 equiv), and PPh3 (6.11 g,
23.3 mmol, 1.1 equiv) in dry THF (50 mL) was added at 0 °C diethyl
azodicarboxylate (DEAD) (3.66 mL, 23.3 mmol, 1.1 equiv), and the
mixture was stirred at 23 °C for 16 h under argon. The resulting
suspension was filtered, and the remanence was washed with cold THF
1
1
(95:5) to a red solid. Yield: 4.45 g (96%). H NMR: δ 1.37 (d, 3H,
(0 °C, 10 mL) and dried. Yield: 6.80 g (83%). H NMR: δ 2.15 (s,
J ) 7.1 Hz), 1.40 (s, 9H), 2.28 (m, 1H), 2.63 (m, 1H), 3.78 (s, 3 H),
3.95 (dd, 1H, J ) 4.3 + 11.3 Hz), 4.06 (d, 1H, J ) 11.3 Hz), 4.47 (p,
1H, J ) 7.5 Hz), 4.71 (t, 1H, J ) 8.2 Hz), 5.14 (bs, 1H), 5.39 (d, 1H,
J ) 7.8 Hz), 6.97 (d, 2H, J ) 8.9 Hz), 7.79 (d, 2H, J ) 8.5 Hz), 7.95
(dd, 4H, J ) 8.5 + 8.9 Hz). 13C NMR: δ 18.33, 28.28, 34.55, 47.79,
52.24, 52.50, 57.73, 75.90, 79.65, 113.40, 115.69, 118.56, 123.14,
125.56, 133.16, 147.31, 154.58, 155.01, 159.77, 171.84, 172.05.
N-Boc-L-Ala-L-trans-Pro(4-{4-[(E)-2-(4-cyanophenyl)-1-diazenyl]-
phenoxy})-NH2 (9). Compound 8 (4.0 g, 7.67 mmol) was dissolved
in a solution of ammonia in methanol (8 M, 200 mL) and stirred for
12 h at 60 °C. The mixture was concentrated in vacuo to a red solid.
3H), 2.55 (m, 2H), 3.75 (s, 3H), 3.91 (m, 2H), 4.80 (bd, 1H, J ) 10
Hz), 5.20 (bs, 1H), 6.97 (d, 2H, J ) 8.9 Hz), 7.82 (d, 2H, J ) 8.6 Hz),
7.97 (d, 4H, J ) 8.7 Hz). 13C NMR: δ 22.28, 34.89, 52.38, 53.00,
57.05, 76.01, 113.50, 115.81, 118.53, 123.14, 125.51, 133.17, 147.31,
154.61, 159.68, 169.43, 171.19.
N-Ac-L-trans-Pro(4-{4-[(E)-2-(4-cyanophenyl)-1-diazenyl]phe-
noxy})-OH (5). To a stirred solution/suspension of 4 (6.62 g, 16.9
mmol) in THF (150 mL) at 0 °C was added LiOH (611 mg, 25.5 mmol,
1.5 equiv) in water (50 mL), and the mixture was stirred for 16 h at 5
°C. The mixture was acidified with 10% aqueous citric acid to pH ∼3
and extracted with chloroform (4 × 200 mL). The combined organic

Azobenzene Peptide for Holographic Information Storage
J. Am. Chem. Soc., Vol. 121, No. 20, 1999 4741
Figure 8. Recording-erasure cycles in a ∼13-µm-thick film of 10
irradiated from (A) the glass side and (B) the film side.
Figure 7. (A) First-order diffraction efficiency as a function of time,
measured during the grating formation in ∼13-µm-thick film of 10.
From the observed diffraction efficiency of 80% and a film thickness
of about 10 µm, an index modulation of approximately 0.08 at 633 nm
can be calculated, assuming that diffraction is due to change in refractive
index alone. (B) Plot of 1/t (reciprocal time to reach 70% diffraction
efficiency) against intensity for 10.
(m, 4H), 4.70 (m, 3H), 5.05 (m, 1H), 5.18 (m, 1H), 6.97 (m, 4H), 7.75
(m, 4H), 7.90 (m, 8H). 13C NMR (CDCl3): δ 17.24, 22.22, 33.80, 33.81,
47.15, 52.33, 53.06, 58.48, 59.27, 75.63, 75.91, 113.22, 115.48, 115.58,
118.44, 123.01, 125.41, 133.04, 146.99, 154.36, 159.75, 159.98, 170.26,
170.60, 171.91, 173.07. MALDI-TOF-MS (DHB): m/z (M + H+) found
(calcd) 768.2 (767.8).
Preparation of DNO Films. Films were prepared by dissolving 25
mg of DNO in a solution of hexafluoro-2-propanol-trifluoroacetic
acid-methylene chloride (76:18:6 (v/v/v), 400 µL). The resulting red
solution was filtered through a syringe filter (0.7 µm pore size). Next,
10-14 drops of the solution were cast on a glass plate (20 mm in
diameter) in a desiccator, and vacuum was immediately applied. The
film was dried for 30 min in the desiccator and then transferred to an
oven at 90 °C overnight. The thickness of the film in question was
measured with a Dektak profiler. Films with a thickness ranging from
1 to 15 µm were prepared to produce the reference curve shown in
Figure 7. The thickness of the individual films examined typically had
a variation in the thickness of less than (3%.
Yield: 3.88 g (>99%). 1H NMR: δ 1.35 (d, 3H, J ) 6.8 Hz), 1.40 (s,
9H), 2.38 (m, 1H), 2.68 (m, 1H), 3.95 (m, 2H), 4.49 (t, 1H, J ) 6.8
Hz), 4.79 (dd, 1H, J ) 6.3 + 7.8 Hz), 5.21 (m, 1H), 5.51 (d, 1H, J )
7.8 Hz), 6.13 (m, 1H), 7.01 (d, 2H, J ) 8.9 Hz), 7.77 (d, 2H, J ) 8.5
Hz), 7.93 (dd, 4H, J ) 8.5 + 8.9 Hz). 13C NMR: δ 18.10, 28.23,
33.39, 47.92, 52.16, 58.35, 75.92, 79.73, 113.26, 115.66, 118.47, 123.04,
125.45, 133.06, 147.13, 154.51, 155.01, 160.05, 172.78, 172.94.
N-Ac-L-trans-Pro(4-{4-[(E)-2-(4-cyanophenyl)-1-diazenyl]phe-
noxy})-L-Ala-L-trans-Pro(4-{4-[(E)-2-(4-cyanophenyl)-1-diazenyl]-
phenoxy})-NH2 (10). Compound 9 (2.0 g, 3.95 mmol) was stirred in
trifluoroacetic acid-methylene chloride (1:1, 100 mL) at 0 °C for 1.5
h. The mixture was concentrated in vacuo and coevaporated with
benzene-methylene chloride (1:1, 3 × 50 mL). The resulting red oil
was dissolved in methylene chloride (200 mL), and Et3N (2.0 mL, 11.8
mmol, 3.0 equiv) was added. A solution of 5 (1.79 g, 4.74 mmol, 1.2
equiv), Et3N (2.0 mL, 11.8 mmol, 3.0 equiv), and benzotriazol-1-
yloxytris(dimethylamino)phosphonium hexafluorophosphate (BOP) (2.09
g, 4.74 mmol, 1.2 equiv) in methylene chloride (200 mL) was added,
and the mixture was stirred for 12 h. The mixture was concentrated in
vacuo to approximately 15 mL and purified by flash chromatography
on a basic alumina column (4 × 3 cm) using chloroform-methanol
(85:15) as eluent. The resulting mixture was concentrated in vacuo to
a red solid, which was dissolved in ethyl acetate (150 mL) and washed
with aqueous KHSO4 (10% w/w, 2 × 100 mL), water (2 × 100 mL),
and aqueous NaHCO3 (2 × 100 mL), dried with MgSO4, and
Results and Discussion
It was envisioned from previous work18 that constraining the
conformational freedom of the DNO backbone structure would
lead to a more efficient alignment of the neighboring chro-
mophores. It appears logical that the coordinated reorientation
of chromophores in a molecule will be more efficient the more
strongly or more rigidly the chromophores are held together.
Based on this rationale, proline-based DNO, shown schemati-
cally in Figure 3, was designed. While retaining a similar
molecular molecular geometry, the proline-based backbone
would be more rigid than the ornithine-based one due to the
pyrrolidine ring, which also mimics the sugar ring of the DNA
1
concentrated to a red solid. Yield: 2.65 g (88%). H NMR (CDCl3):
(26) Merrifield, R. B. J. Am. Chem. Soc. 1963, 85, 2149-2154.
(27) Merrifield, B. Science 1986, 232, 341-347.
δ 1.40 (d, 3H, J ) 6.4 Hz), 2.06 (s, 3H), 2.4-2.7 (m, 4H), 3.87-4.05

4742 J. Am. Chem. Soc., Vol. 121, No. 20, 1999
Rasmussen et al.
Scheme 1
after the two beams from the argon ion laser were switched
off, and, after 3 s, one of the argon ion beams (12.5 mW/mm2)
was switched on again. The diffraction efficiency decreased
rapidly and reached a level of 1.2% in less than 3 s. It is noted
that the erasure process is much faster than that previously
observed with ornithine-based DNO. The duration of the
exposure required to reach a certain diffraction efficiency is
inversely proportional to the intensity of the recording beams
over a wide range, indicating that no significant thermal effects
are involved in the recording process. Thus, a plot of the
reciprocal of the time it takes to reach 70% diffraction efficiency
against the intensity of the laser power gave a straight line
(Figure 7B). For holograms recorded in 10, with an exposure
time of about 1 s, lifetimes of now more than 1 year have been
observed. The holograms can be erased thermally if heated to
about 110 °C. No first-order thermal transitions were detected,
by differential scanning calorimetry, in the temperature range
from 25 °C to the decomposition temperature (about 220 °C).
Figure 9. (A) Maximum first-order diffraction efficiency obtained
(after ∼1 s) as a function of film thickness. (B) Absorbance at 510 nm
as a fucntion of film thickness.
backbone structure. A series of proline-based DNOs was
prepared by Merrifield solid-phase synthesis,26,27 and their
recording properties were examined by polarization hologra-
phy.28 Films of good optical quality with a thickness of
approximately 10 µm were obtained from hexafluoro-2-pro-
panolic solutions. When the films were examined in a polariza-
tion microscope, they exhibited no birefringence. The setup used
for conventional two-beam polarization holography is shown
in Figure 5.29 Two orthogonally circularly polarized beams at
488 nm from an argon ion laser, with a total intensity of 25
mW/mm2, were used to record holographic gratings in the film.
The spacing of the gratings is about 3.7 µm, and the spot size
is about 3 mm in diameter. A circularly polarized beam at 633
nm from a 4.2-mW helium-neon laser was used for read-out.
DNOs were tested, with the best one identified from this series
as dimer 10 (Chart 1).30 The UV-visible absorption spectrum
of dimer 10 is shown in Figure 6, and as shown in Figure 7A,
it gave a first-order diffraction efficiency of 75% in about 800
ms. The maximum efficiency, approximately 80%, was reached
in about 1 s. At recording times longer than 1 s, the efficiency
is lowered. The response of 10 is more than 350 times faster
than that of the ornithine-based dimer. After exposure, the
illuminated film, when examined between crossed polarizers,
exhibits birefringence. The diffraction efficiency remained stable
o
However, a glass transition was observed at around 116 C,
suggesting that the material is amorphous.
As shown in Figure 8A, recording-erasure cycles can be
performed in a film of 10 if the film-coated glass is irradiated
through the glass side. If the film was irradiated directly, that
is, from the film side, the first-order diffraction efficiency
decreased rapidly after a few cycles (see Figure 8B) as the
diffraction was spread into other orders. This is due to a large
surface relief, which is created when the film is irradiated from
the film side, as seen by using atomic force microscopy (not
shown). An obvious explanation of the surface relief is to be
found in the role played by the free surface. When the laser
beams enter from air into the film, the first surface is the free
surface of the film. Due to the higher intensity, a surface relief
is likely to occur at this surface. On the other hand, when the
laser beams enter the film through the glass surface, the first
film surface is the surface in contact with the glass. No surface
relief is possible at this interface. When the beams leave the
free surface of the film, their intensity has decreased due to the
absorption in the film. The surface relief has thus an extremely
low magnitude. Studies are currently underway to examine this
phenomenon. In Figure 9A, it is seen that the first-order
diffraction efficiency decreases rapidly if the film thickness
becomes less than 5 µm. The reference curve shown in Figure
9B can be used for indirectly estimating the thickness of a film.
Another important factor for the development of holographic
recording materials is the ease of their preparation for large-
(28) It is not necessary to use polarization holography for these materials.
It is possible to record with the same polarization. However, the diffraction
efficiency varies according to the polarization combination. It is possible
to achieve 100% diffraction efficiency with two orthogonally circularly
polarized beams for the writing and one circularly polarized beam for the
read out, even in thin films. See, e.g.: Todorov, T.; Nikolova, L.; Tomova,
N. Appl. Opt. 1984, 23, 1342.
(29) A general discussion on interference, diffraction, and holography
can be found in the following: Collier, R. J.; Buckhardt, C. B.; Lin, L. H.
Optical Holography; Academic Press: New York, 1971.
(30) The synthesis and recording properties of the complete series will
be reported elsewhere.

Azobenzene Peptide for Holographic Information Storage
J. Am. Chem. Soc., Vol. 121, No. 20, 1999 4743
Scheme 2
in the pyrrolidine ring, with an azobenzene chromophore in the
(R)-configuration using the Mitsunobu reaction.31 Synthesis of
the two building blocks 5 (Scheme 1) and 9 (Scheme 2) was
readily accomplished in three and four steps, respectively, and
in 76% and 79% overall yields, respectively. After deproctection
of the Boc group, 9 was coupled to 5, providing 10 in 88%
1
yield. The peptide was characterized by H and 13C nuclear
magnetic resonance spectroscopy and shown to have the
expected molecular mass by matrix-assisted laser desorption/
ionization time-of-flight mass spectrometry. Small amounts of
impurities could be detected by high-perfomance liquid chro-
matography (HPLC). A small fraction of 10 was purified by
HPLC, but this did not lead to any improvement of the recording
properties, indicating that the behavior of 10 is not affected
significantly by the presence of impurities. Since 10 contains
proline residues, studies are in progress to investigate the
possible cis-trans isomerization of the peptide bonds.
Conclusion
In summary, a strategy that focuses on increasing the rigidity
of the backbone has led to the design of a new class of
azobenzene peptides that eliminates some of the limitations
associated with the first generation of DNOs. With regard to
the response time, by adding further structural constraints to
the molecule, there may be a good deal of room for further
improvement. For potential use in volume holography, recording
in thick films of the order of 1 mm is needed.32 The greatly
improved solubility properties of 10 should increase the chance
of finding a polymer host more suitable for “dilution” of DNO
than the previously reported one consisting of polystyrene-
grafted polyethylene.18,33 Besides peptides, the work described
here suggests that a wide range of other small and rationally
designed, rigid azobenzene molecules may prove valuable for
holographic storage of information. Widespread applications of
any azobenzene system will rely on the commercial production
of blue laser diodes, which is expected to be realized in the
very near future.34
scale availability. Because of their insolubility in organic
solvents other than trifluoroacetic acid and hexafluoro-2-
propanol, the ornithine-based DNOs could only be assembled
by solid-phase synthesis, which is difficult to scale-up to
multigram or larger quantities. By contrast, the proline-based
ones are soluble in methylene chloride, chloroform, acetonitrile,
and other solvents compatible with solution-phase synthesis.
This unexpected property greatly facilitates large-scale fabrica-
tion, which is a prerequisite for the production of low-cost
materials. A solution-phase synthetic strategy was developed,
and 10 was synthesized on a 3-g scale from commercially
available hydroxyproline derivatives through the procedure
outlined below. A key step involves stereospecific substitution
of the hydroxy group of 2 and 6, which has the (S)-configuration
Acknowledgment. This paper is dedicated to Professor
James P. Tam on the occasion of his 50th birthday. We thank
Dr. Walther Batsberg Pedersen for his expert assistance in
chromatography. We thank Ms. Anne Bønke Nielsen and Ms.
Britt J. Harder for skilled technical assistance. This work was
supported by the Danish Natural Science Research Council and
the Danish Materials Technology Development Program, and
by a Ph.D. grant from the Danish Research Academy (P.H.R.).
Supporting Information Available: Details of the solution-
phase synthesis of dimer 10 (PDF). This material is available
free of charge via the Internet at http://pubs.acs.org.
(31) Mitsunobu, O. Synthesis 1981, 1-28.
(32) Peyghambarian, N.; Kippelen, B. Nature 1996, 383, 505.
(33) Berg, R. H.; Almdal, K.; Batsberg Pedersen, W.; Holm, A.; Tam,
J. P.; Merrifield, R. B. J. Am. Chem. Soc. 1989, 111, 8024-8026.
(34) Rigby, P. Nature 1996, 384, 610.
JA981402Y