Synthesis and nonlinear-optical properties of vinyl-addition poly(norbornene)s

Macromolecules 2004, 37, 5163-5178

5163

Synthesis and Nonlinear-Optical Properties of Vinyl-Addition

Poly(norbornene)s

Ki Hon g P a r k ,^†,‡Rober t J . Tw ieg,*^,†R. Ra vik ir a n ,^§La r r y F . Rh od es,*^,§

Rober t A. Sh ick ,^§Diego Ya n k elevich , a n d An d r e Kn oesen

Chemistry Department, Kent State University, Kent, Ohio 44240, Promerus LLC,

9921 Brecksville Road, Brecksville, Ohio 44141, and Department of Electrical and Computer

Engineering, University of California at Davis, Davis, California 95616

Received February 23, 2004; Revised Manuscript Received April 27, 2004

ABSTRACT: Vinyl-addition poly(norbornene) copolymers functionalized with nonlinear optical chro-

mophore side groups have been prepared using (η⁶-toluene)Ni(C₆F₅)₂, and their electrooptic properties

have been characterized. The nickel complex used to polymerize the norbornene monomers is tolerant to

many functional groups found in nonlinear optical chromophores although nitriles and amines other than

trisubstituted amines strongly inhibit the reaction. A vinyl-addition copolymer of hexylnorbornene and

a norbornene-functionalized Disperse Red 1 chromophore was scaled up and studied in detail. Initial

studies indicate that electric field poling is effective but that relaxation of polar order in the

poly(norbornene) is faster than in a comparable methacrylate copolymer.

In tr od u ction

and host polymers in numerous combinations have been

examined with very mixed results. To date, there still

seems to be no general consensus on any unique design

strategy that leads to the simultaneous optimization of

all the required and often seemingly incompatible

physical properties demanded of a single clearly supe-

rior electrooptic material. This should probably be no

surprise given the inherent complexity of the problem

at hand and also given the diverse options that organic

chemistry and polymer science present as potential

solutions. The matrix of possibilities is huge. Consider

first that the “magic bullet” chromophore (with a large

molecular µâ product, miscibility, and other properties

leading to creation of sufficient nonlinear susceptibility

ø⁽²⁾manifested as a large electrooptic coefficient r₃₃, a

small V_π, and last but never least, sufficient photo-

chemical and thermochemical stability) remains to be

identified. Next, consider the myriad of permutations

on the polymer class to select from (acrylate, polyimide,

polyquinoline, sol-gel, epoxy, etc.) and then, finally, the

exact relationship between the chromophore and poly-

mer (guest-host, covalent main-chain, covalent side-

chain, linear, dendrimer, nonfunctionalized and nonre-

active, functionalized and reactive, etc.). The choice of

the appropriate NLO polymer is one beset by many

tradeoffs. On one hand the relative ease and compat-

ibility in processing polymers is promoted as a great

asset. On the other hand, this enhanced processability

tends to compromise (but not negate) the achievement

of large and stable polar order. Therefore, one critical

issue in NLO polymeric material has been the enhance-

ment of long-term stability of NLO properties. Unlike

inorganic materials, where the polar order is “built in”,

in the case of polymers the NLO chromophores (the

donor-bridge-acceptor unit) must be oriented (poled)

by an external electric field and this polar order must

be retained even at elevated temperatures in the

absence of that poling field. Therefore, it is very

important to sustain the polar alignment of NLO

chromophores in the polymer matrix for a long period.

For this reason, high glass temperature (T_g) polymers

such as polyimides have been proposed as suitable NLO

Information processing and telecommunications tech-

nologies are approaching a barrier to meet current much

less anticipated demands of signal propagation and

switching speeds available based solely on electronics.

As a result, the future ultrafast computing and com-

munications systems will inevitably integrate photonics

on a much larger scale. For such optical communication

networks, one of the critical components required is the

optical fiber by which data is relayed. Another critical

component is an optical modulator or switch required

to put the information on and off the fiber and to route

the information. In existing optical networks, these

components rely heavily on silica to provide the infor-

mation transport function and on lithium niobate to

provide the information switching and routing function,

although numerous other schemes and materials have

also been proposed and studied. In the case of the

switching and routing function, organic and polymer

electrooptical media that can emulate lithium niobate

(and eventually have performance superior to lithium

niobate) have been the most intensively investigated.¹

This sustained study has been motivated by the convic-

tion that these polymers can provide a cost-effective

solution due to their intrinsic design and processing

flexibility. Furthermore, in addition to applications in

computing and communications, such polymeric non-

linear optical (NLO) media may find additional utility

in photorefractive systems, phase array radar, and

many other related applications.

Over the past 2 decades significant progress has been

made, leading to the design and fabrication of more

practical and reliable organic and polymeric photonics

materials and devices. Even in the area of electrooptics

alone, an enormous variety of nonlinear chromophores

* To whom correspondence should be addressed. E-mail:

larry.rhodes@promerus.com.

^†Kent State University.

^‡Current address: KIST, Electronic Materials

Research Center, Seoul, 130-650, Korea.

^§Promerus LLC.

& Devices

University of California at Davis.

Published on Web 06/16/2004

5164 Park et al.

Macromolecules, Vol. 37, No. 14, 2004

toluene)Ni(C₆F₅)₂to give the chromophore-functional-

ized poly(norbornene)s. We also report preliminary

nonlinear optical measurements on one of the new NLO

polymers we have prepared.

Exp er im en ta l Section

Ma ter ia ls. Starting materials for chromophore synthesis

including 4-aminobenzonitrile (1), 2-(N-ethylanilino)ethanol

(2), 4-nitrophenylacetic acid (4), 6-chloro-1-hexanol (6), 4-N,N-

(diethylamino)salicylaldehyde (7), ethyl cyanoacetate (9), iso-

phorone (10), 4-N,N-dimethylaminobenzaldehyde (12), and

Disperse Red 1 (C1) were purchased from the Aldrich Chemi-

cal Co. The 1,5,5-trimethylcyclohex(2-enylidene)malononitrile

(5) was provided by IBM Almaden Research Laboratory for

research use. The dibutyltin oxide (98%) was purchased from

Acros Organics. The 5-norbornene-2-ethyl ester (13), 5-nor-

bornene-2-methanol (14), and 5-norbornene-2-hexyl monomers

(18) were provided by the BFGoodrich Co. Anhydrous toluene

(99.8%) used as a polymerization solvent was purchased from

Aldrich. (η⁶-toluene)Ni(C₆F₅)₂was prepared according to a

literature method.¹⁵The DR1-PMMA copolymer used in this

study for comparative purposes was obtained from the IBM

Almaden Research Laboratory.¹⁶

F igu r e 1. Ring opening metathesis type vs vinyl-addition type

polymerization.

polymer backbones. However, the large birefringence

associated with many polyimides make them as a class

of materials less than ideal for electrooptics. As an

alternative, cross-linking strategies have been examined

to freeze in polar order by creation of new chemical

bonds that might better lock chromophores in a polar

arrangement.²

A class of polymers that exhibit very good optical

properties including low birefringence are poly(cyclic

olefins). Ring-opening polymerization of cyclic olefins,

such as norbornene, give ROMP-type polymers (Figure

1). However, these types of polymers exhibit T_g’s that

are significantly lower (typically below 200 °C) than

polyimides and they suffer from thermooxidative insta-

bility due to the presence of unsaturation in the

backbone. Despite these potential problems, there have

been a few brief reports about application of ROMP

systems as nonlinear optical polymers (NLOP)^3,4and

as other kinds of optical and electronic materials such

as side chain liquid crystals (SCLCP),⁵electrolumines-

cent materials (EL),⁶and organic light emitting diode

(OLED).⁷

Another method of preparing cyclic olefin polymers

is by vinyl addition polymerization (Figure 1). Vinyl

addition polymers of norbornene were originally re-

ported as early as 1963.^8aA report of vinyl addition

polymerization of functional norbornenes appeared in

1967.^8bSince then there have been a number of reports

describing early transition metal metallocene and late

transition metal catalysts for their preparation.⁹The

vinyl-polymerized norbornenes synthesized using cer-

tain metallocenes were found to be highly tactic, high

melting, semicrystalline materials that would seem to

be unsuitable for optical applications.¹⁰Those polymers

produced by late transition metals were found to be less

stereoregular. Recently, new group VIII transition metal

catalyst systems for the vinyl addition polymerization

of norbornene-type monomers have appeared.¹¹These

addition-type poly(norbornene)s have high T_g’s in the

range of about 180-370 °C depending on the type of

pendant groups (R) attached to the norbornene, whereas

the highly unsaturated ring-opening metathesis polym-

erization (ROMP) polymer of the parent norbornene has

a T_gof only about 35 °C.¹²Furthermore, addition-type

poly(norbornene)s are amorphous, exhibit low birefrin-

gence and can exhibit low moisture absorption.¹³They

are also more thermally stable than the corresponding

ROMP polymers since they contain no unsaturation.

Recently, photochromic poly(norbornene)s have been

successfully synthesized by this vinyl addition polym-

erization method.¹⁴

Mea su r em en ts. Ch em ica l Ch a r a cter iza tion . Routine ¹H

NMR spectra were obtained on a Bruker AMX300 NMR (300

MHz) spectrometer while a Varian Inova 500 MHz spectrom-

eter MHz spectrometer was used for ¹H NMR examination of

the polymer samples. The maximum absorption wavelengths

(λ_max) of compounds were determined by HP 8453 UV-visible

spectrophotometer in acetonitrile (CH₃CN). The melting points

(peak points and onset points) of all synthesized compounds

were measured on TA Instruments 2920 DSC at a heating rate

of 5 °C/min under nitrogen. Molecular weight measurements

were determined by gel-permeation chromatography (GPC)

using a refractive index detector. Polymer samples were

dissolved in stabilized THF (50 mg/20 mL) and filtered through

a 0.20 µm Teflon filter. All the molecular weights are measured

relative to polystyrene standards. Dynamic mechanical analy-

sis (DMA) experiments were done using a Rheometrics RSA

II analyzer as per the ASTM D5026-95a method (DMA of

plastics in tension). The polymer films were heated in 3° C

steps with the dwell time at each temperature being 1 min.

Polymer samples for DMA were prepared by casting a solution

of the polymer (in mesitylene) onto a glass plate and allowing

the solvent to evaporate overnight. The films were then baked

at 200 °C under vacuum to ensure complete removal of solvent.

Non lin ea r Op tica l Ch a r a cter iza tion . Thin films were

prepared by spin casting polymer solutions onto a soda-lime

glass substrate followed by baking to remove residual solvent.

The film thickness was measured by contact profilometry. The

second harmonic generation (SHG) measurements were per-

formed using a p-polarized Q-switched Nd:YLF laser beam

incident at Brewster’s angle on the surface coated with the

electrooptic polymer. The laser operates at a wavelength of

1057 nm with a 1 kHz repetition rate. The generated second

harmonic and fundamental beams are filtered using dichroic

mirrors and filters. The second harmonic signal is detected

with a PMT and lock-in amplifier. The average fundamental

power was kept constant throughout the measurements as

verified with a pyroelectric power meter. The refractive index

as a function of wavelength was measured by variable angle

spectroscopic ellipsometry.

Assuming a simple scalar model, the macroscopic dielectric

polarizability is

P ≈ ꢀ_oø⁽¹⁾E + ꢀ_oø⁽²⁾E²+ ꢀ_oø⁽³⁾E³

This investigation describes our preliminary attempts

to exploit vinyl-addition polymerized poly(norbornene)

as a new polymer class for NLO applications. Here we

demonstrate the preparation of norbornene monomers

functionalized with some conventional NLO chro-

mophores and their successful polymerization with (η⁶-

where ø⁽¹⁾, ø⁽²⁾, and ø⁽³⁾are the macroscopic linear polarizability,

second-order nonlinear polarizability, and the third-order

nonlinear polarizability, respectively. If the applied electric

field consists of a strong static electric field E₀and an optical

beam E₁cos(ωt), then substituting E ) E₀+ E₁cos(ωt)

into the expression for polarizability and extracting the

Macromolecules, Vol. 37, No. 14, 2004

Vinyl-Addition Poly(norbornene)s 5165

second harmonic term gives

of an EFISH contribution will lead to erroneous conclusions.

To eliminate the EFISH contribution, the surface charge is

removed by immersing the samples into unpurified water for

5-10 min. To illustrate the importance of eliminating the

EFISH contribution in this manner, we corona poled a LCP 1

film and performed measurements of the surface potential and

second harmonic generated by the film after poling and after

immersing the sample in water. The surface potential is

measured with an electrostatic voltmeter that permits voltage

measurements of electrostatic sources without physical con-

tact. A film of LCP 1 (480 nm thick) cast on a 1 mm thick soda-

lime glass substrate was used for the experiment. Immediately

after poling, the surface potential, measured over an area of

5 × 2 cm, was in the 165 ( 15 V range resulting in an average

electric field of 344 V/µm. The film generated a second

harmonic signal of 8.6 au. After the sample is immersed in

water, the electric field is reduced to 46 V/µm and the second

harmonic signal is reduced to 2.5 au. Given such a relatively

small electric field, the assumption is made that the EFISH

contribution is negligibly small, and second harmonic signal

1

2

3

2

P

^2ω) ꢀ_oE₁ø⁽²⁾

+

ø⁽³⁾E₀cos(2ωt) ) ꢀ_oE₁ø_eff⁽²⁾cos(2ωt)

(

)

(1)

The dominant macroscopic second-order polarization element

is

ø⁽²⁾) Nâ cos³θ >

where N is the chromophore volume density, â is the dominant

second-order microscopic polarizability, and cos³θ is an order

parameter that can be expressed in terms Langevin function

of order 3.¹⁷A noncentrosymmetric alignment as induced by

electric field poling induces a ø⁽²⁾in the polymer film. In this

investigation a corona discharge is used to apply a large elec-

tric field to the polymer film to orient the nonlinear chro-

mophore.¹⁸The intensity of the generated second harmonic is

is dominated by the contribution from ø⁽²⁾

.

I_2ω≈ Ad²ø_eff⁽²⁾²I_ω

2

Measurements of the new materials, such as LCP 1, were

done in comparison with a well-known and well-characterized

NLO polymer DR1-MMA. The optimal poling conditions for

the DR1-MMA films, defined as repeatable second harmonic

intensity signals during and after poling measured on multiple

samples prepared in similar manner, were established at ∼100

°C with a voltage around 5.7 kV and current of about ∼2 µΑ

(note that these conditions are specific to our poling configu-

ration). The ratio of effective susceptibilities of the reference

material with respect to the material under test is calculated

by the equation

where A is a proportionality constant and d is the film

thickness provided that the film is considered to be lossless

and the thickness is small so that |n₂- n₁|d < λ_2ωwhere n₂

and n₁are the indices of refraction at the second harmonic

wavelength λ_2ωand fundamental wavelength λ_ω, respectively.

The corona poling was performed in either a poling station

or in an in situ cell that provides simultaneous poling and SHG

monitoring. In the poling station, the substrate is placed on a

temperature-controlled grounded aluminum block. Typical

conditions in the poling station for 1 mm thick sodalime

substrates with a corona wire suspended 1 cm above the

substrate involve poling voltages in the range of 6 kV and

poling currents in the range of 2 µA. Poling is performed at

(2)

2ω

ω

2

ø_REFeff

P_REF

P_TESTL_TEST

)

(2)

2ω

ω

(2)

(

)

P

P_REFL_REF

ø_TESTeff

x

TEST

1

an elevated voltage and temperature for at least /₂h and then

gradually cooled to room temperature while the poling field is

still applied. Use of a temperature-controlled in situ poling

cell permits more control on the optimization of the poling

process. For such measurements the glass substrate has an

indium tin oxide coating on the opposite surface on which the

polymer is deposited. A high-voltage current-controlled power

supply is connected to a thin tungsten wire (40 µm diameter)

held at 1 cm above the polymer film and the ITO-coated side

is grounded. A thermocouple is placed on the edge of surface

of the polymer film and it is electrically isolated from the

discharge (e.g., by wrapping it in Teflon tape) to avoid current

flow from the corona discharge to the temperature controller

through the thermocouple. To make sure that the current from

the corona wire is conducted through the polymer film and

substrate to ground, it is measured between the power supply

and the corona wire and between the conducting plane in

contact with the glass substrate and ground. A measurement

of the temperature distribution on the sample surface is used

to account for the difference, in the in situ SHG measurements,

between the region where the second harmonic measurement

is performed and the edge of the sample where the tempera-

ture is measured. The lock-in amplifier, temperature control-

ler, and one of the picoamperimeters were interfaced to a

computer to record the second harmonic signal, temperature,

and poling current as a function of time. All films were

maintained at elevated voltage and temperature for at least

30 min and then cooled to room temperature while keeping

the poling voltage constant. After corona poling, a significant

surface charge remains on the polymer surface that must be

removed to obtain an accurate assessment of the second

harmonic signal produced by solely by the orientation of the

nonlinear chromophores. As can be seen in eq 1 the presence

of a static electric field will enhance the second harmonic

intensity through the third-order optical nonlinearity ø⁽³⁾. This

contribution is called the electric field enhanced second

harmonic (EFISH). If, for example, after poling the change in

the orientation of the nonlinear chromophore is of interest,

since the surface charge will also slowly decay, the presence

where P^ωis the fundamental laser power incident on the film,

P^2ωis the second harmonic power generated by the film, and

L is the thickness of a particular film.

Syn th esis. 4-[(2-Hyd r oxyeth yl)(m eth yl)a m in o]ben za l-

d eh yd e (3). Into a 500 mL flask containing a magnetic stirring

bar and 250 mL of dimethyl sulfoxide were added 30.0 g (0.40

mol) of 2-(methylamino)ethanol, 52.0 g (0.38 mol) of potassium

carbonate, and 5 drops of tricaprylmethylammonium chloride.

The mixture was heated to 90 °C with stirring, and then 37.2

g (0.30 mol) of 4-fluorobenzaldehyde was added dropwise. The

reaction was continued for 35 h at 90 °C, and after cooling

the mixture was poured into 1 L of ice water with vigorous

stirring. The product was extracted with dichloromethane, and

the organic layer was separated, dried with magnesium

sulfate, and evaporated to give a liquid that was added

dropwise to cold petroleum ether with stirring. The precipi-

tated yellow solid was collected and recrystallized from toluene

twice to give light yellow fine crystals of 3. Yield: 31 g (58%).

19

1

DSC: mp 71.0 °C (onset 69.8 °C). H NMR (CDCl₃): δ 9.71

(s, 1H, CHO), 7.65 (d, 2H, ArH), 6.77 (d, 2H, ArH), 3.83 (t,

2H, CH₂), 3.60 (t, 2H, CH₂), and 3.12 (s, 3H, NCH₃).

4-(Diet h yla m in o)-2-[(6-h yd r oxyh exyl)oxy]b en za ld e-

h yd e (8). Into a 250 mL flask were placed 9.66 g (0.05 mol) of

4-(diethylamino)salicylaldehyde (7), 13.66 g (0.10 mol) of

6-chloro-1-hexanol (6), and 100 mL of N,N-dimethylformamide

(DMF) and 30 mL of toluene. Anhydrous potassium carbonate,

3.50 g (0.025 mol), was added to the solution, and the mixture

was heated for 24 h while distilling out water by means of a

Dean-Stark trap. After cooling, the product was isolated by

extraction with ethyl acetate vs water and the organic layer

was dried with magnesium sulfate and condensed by rotary

evaporation. The residual liquid was chromatographed on a

silica gel column (ethyl acetate/petroleum ether ) 1/1), to give

the product 8 as dark brown oil. Yield: 9.70 g (66%). ¹H NMR

(CDCl₃): δ 7.72 (d, 1H, ArH), 6.30 (d, 1H, ArH), 6.07 (s, 1H,

ArH), 4.04 (t, 2H, OCH₂), 3.67 (t, 2H, CH₂OH), 3.42 (q, 2H,

NCH₂), 1.86-1.85 (m, 2H, CH₂), 1.62-1.45 (m, 4H, CH₂), and

1.22 (t, 6H, CH₃).

5166 Park et al.

Macromolecules, Vol. 37, No. 14, 2004

Eth yl 2-Cya n o-2-(3,5,5-tr im eth yl-2-cycloh exen ylid en )-

a ceta te (11). To a 500 mL flask equipped with a Dean-Stark

trap and a condenser were added 27.64 g (0.2 mol) of isophor-

one (10), 22.60 g (0.20 mol) of ethyl cyanoacetate (9), 200 mL

of benzene, 16 mL of acetic acid, and 8 g of sodium acetate.

The mixture was stirred at reflux while distilling benzene for

28 h. After cooling, water was added and the mixture was

transferred to a separatory funnel; the organic layer was

separated, dried with magnesium sulfate, filtered, and then

evaporated. The resulting viscous oil was transferred to a 100

mL one-neck flask and distilled under vacuum (bp 168-170

°C/6 Torr) to give the product 11 as a light yellow liquid.

min): mp 144.3 °C, onset 142.4 °C. ¹H NMR (CDCl₃): δ 7.95-

7.90 (m, 4H, ArH), 7.75 (d, 2H, ArH), 6.87 (d, 2H, ArH), 3.91

(t, 2H, OCH₂), 3.69-3.55 (m, 4H, CH₂NCH₂), and 1.27 (t, 3H,

CH₃).

2-Meth yl-4-[(E)-2-(4-n itr op h en yl)-1-eth en yl]a n ilin o-1-

eth a n ol (C3). Compound 3 (8.96 g, 0.05 mol), 18.11 g (0.1 mol)

of 4-nitrophenylacetic acid (4) and 30 mL of DMF were placed

into a 100 mL flask. After the reaction was warmed to 40 °C,

piperidine (10 mL) was added dropwise, and the mixture was

heated at 100 °C for 10 h. The cooled reaction mixture was

poured into water to precipitate the product that was isolated

by suction filtration and recrystallized from ethanol, giving

C3 as red crystals. Yield: 3.73 g (25%). UV (CH₃CN): λ_max

437 nm. DSC (5 °C/min): mp 184.7 °C, onset 179.0 °C. ¹H NMR

(CDCl₃): δ 8.18 (d, 2H, ArH), 7.56 (d, 2H, ArH), 7.45 (d, 2H,

ArH), 7.20 (d, 1H, CHd), 6.94 (d, 1H, CHd), 6.81 (d, 2H, ArH),

3.86 (t, 2H, OCH₂), 3.56 (t, 2H, NCH₂), and 3.06 (s, 3H, NCH₃).

2-[3-((E)-2-{4-[(2-Hydr oxyeth yl)(m eth yl)am in o]ph en yl}-

1-et h en yl)-5,5-d im et h yl-2-cycloh exen ylid en ]m a lon on i-

tr ile (C4). To a 250 mL flask with a magnetic stirring bar

and a condenser were added 3.58 g (0.02 mol) of compound 3,

50 mL of 1-propanol, 3.73 g (0.02 mol) of 1,5,5-trimethylcy-

clohex-(2-enylidene)-malononitrile (5) and 10 drops of piperi-

dine, and then the solution was stirred at 100 °C for 24 h.

After TLC indicated that all of 3 was consumed, the reaction

mixture was cooled to room temperature, was concentrated

by rotary evaporation, and then was poured into petroleum

ether with vigorous stirring. The precipitated material was

collected and recrystallized from toluene to give C4 as fine

red crystals. Yield: 4.90 g (65%). UV (CH₃CN): λ_max501 nm.

DSC (5 °C/min): mp 151.7 (onset 146.3) and 158.8 °C. ¹H NMR

(THF-d₈): δ 7.44 (d, 2H, ArH), 7.13 (d, 1H, CHd), 6.97 (d, 1H,

dCH), 6.74-6.72 (m, 3H, ArH), 3.67 (d, 2H, OCH₂), 3.50 (t,

2H, NCH₂), 3.06 (s, 3H, NCH₃), 2.57 (s, 2H, cyclic CH₂), 2.51

(s, 2H, cyclic CH₂), and 1.05 (s, 6H, (CH₃)₂).

1

Yield: 38.6 g (83%). H NMR (CDCl₃): δ 7.61 (s, 0.5H, dCH),

6.67 (s, 0.5H, dCH), 4.28-4.25 (m, 2H, CH₂CH₃), 2.90 (s, 2H,

cyclic CH₂), 2.55 (s, 2H, cyclic CH₂), 2.12-1.98 (m, 3H, CH₃),

1.35 (t, 3H, CH₂CH₃), and 1.00 (d, 6H, (CH₃)₂).

2-(Eth yla n ilin o)eth yl Bicyclo[2.2.1]h ep t-5-en e-2-ca r -

boxyla te (15). Into a 500 mL flask were placed 66.50 g (0.40

mol) of 5-norbornene-2-carboxylic acid ethyl ester (13), 33.05

g (0.20 mol) of 2-(N-ethylanilino)ethanol (2), and 200 mL of

toluene. Dibutyltin oxide 1.24 g (5 mmol) was added to this

solution, which was heated to 160 °C while continuously

removing toluene over 6 h. The residual mixture was trans-

ferred to a 250 mL flask and vacuum distilled (bp 174-175

°C/0.4 Torr) to give the product 15 as a viscous liquid. Yield:

40.0 g (70%). ¹H NMR (CDCl₃): δ 7.24 (d, 2H, ArH), 6.73 (broad

s, 3H, ArH), 6.20-5.90 (m, 2H, CHdCH), 4.19 (t, 2H, OCH₂),

3.57-3.38 (m, 4H, CH₂NCH₂), 3.19-1.26 (m, 7H, norbornene-

7H), and 1.18 (t, 3H, CH₃).

2-(E t h yl-4-for m yla n ilin o)et h yl Bicyclo[2.2.1]h ep t -5-

en e-2-ca r boxyla te (16). Into a flask containing 100 mL of

DMF was added 30.7 g (0.20 mol) of phosphorus oxychloride

(POCl₃) dropwise at 0 °C, and then compound 15 (28.5 g, 0.10

mol) dissolved in 50 mL of DMF was added slowly using a

dropping funnel at room temperature. The reaction was stirred

at room temperature for 3 h, and then the mixture was

warmed at 90 °C for 1 h. After cooling, the mixture was poured

slowly into ice water with stirring, and aqueous NaOH solution

was added slowly until pH 6 was reached. The mixture was

then extracted with ethyl acetate and the product was purified

by column chromatography (ethyl acetate/petroleum ether )

1/2) to give the product 16 as a viscous yellow liquid. Yield:

6-{5-(Dieth yla m in o)-2-[(E)-2-(4-n itr op h en yl)-1-eth en yl]-

p h en oxy}-1-h exa n ol (C5). Compound 8 (5.87 g, 0.02 mol)

and 4-nitrophenylacetic acid (7.24 g, 0.04 mol) were placed in

a 100 mL flask and dissolved in 10 mL of DMF at 40 °C.

Piperidine (4 mL) was added dropwise into the solution, and

the mixture was heated at 100 °C for 10 h and after cooling

the reaction mixture was poured into water to precipitate the

product. The crude product was recrystallized from ethanol,

giving C5 as red crystals. Yield: 2.10 g (25%). UV (CH₃CN):

1

17.5 g (56%). H NMR (CDCl₃): δ 9.75 (s, 1H, CHO), 7.73 (d,

2H, ArH), 6.80 (d, 2H, ArH), 6.18-5.86 (m, 2H, CHdCH), 4.22

(t, 2H, OCH₂), 3.67-3.47 (m, 4H, CH₂NCH₂), 3.19-1.38 (m,

7H, norbornene-H), and 1.23 (t, 3H, CH₃).

1

λ_max457 nm. DSC (5 °C/min): mp 88.7 °C, onset 85.7 °C. H

NMR (CDCl₃): δ 8.15 (d, 2H, ArH), 7.59-7.46 (m, 3H, ArH),

7.35 (d, 1H, CHd), 6.99 (d, 1H, CHd), 6.31 (d, 1H, ArH), 6.16

(s, 1H, ArH), 4.05 (t, 2H, OCH₂), 3.67 (t, 2H, CH₂OH), 3.39 (q,

4H, NCH₂), 1.92 (t, 2H, CH₂), 1.66-1.48 (m, 6H, CH₂), and

1.21 (t, 6H, CH₃).

2-[4-F or m yl(m eth yl)a n ilin o]eth yl Bicyclo[2.2.1]h ep t-

5-en e-2-ca r boxyla te (17). Into a 100 mL flask were added

4.99 g (0.03 mol) of 5-norbornene-2-carboxylic acid ethyl ester

(13), 3.58 g (0.02 mol) of compound 3, 50 mL of toluene, and

dibutyltin oxide (0.124 g (0.5 mmol)). The solution was heated

at 160 °C while toluene was slowly removed over 5 h using a

Dean-Stark trap. After concentration, the remaining liquid

was purified by column chromatography (ethyl acetate/

petroleum ether ) 1/3), yielding 17 as a yellow viscous oil. The

yield was 4.90 g (82%). ¹H NMR (CDCl₃): δ 9.75 (s, 1H, CHO),

7.74 (d, 2H, ArH), 6.78 (d, 2H, ArH), 6.16-5.81 (m, 2H, CHd

CH), 4.23 (t, 2H, OCH₂), 3.68 (t, 2H, NCH₂), 3.10 (s, 3H, NCH₃),

and 2.89-1.22 (m, 7H, norbornene-H).

Syn th esis of NLO Ch r om op h or es. 4-((E)-2-{4-[Eth yl-

(2-h yd r oxye t h yl)a m in o]p h e n yl}-1-d ia ze n yl)b e n zon i-

tr ile (C2). Into a 100 mL flat-bottom flask were added 20 mL

of concentrated hydrochloric acid and 2.36 g (0.02 mol) of

4-aminobenzonitrile (1) with stirring. In another test tube, 1.38

g (0.02 mol) of sodium nitrite was dissolved in 2 mL of water,

and this aqueous solution was then added dropwise into the

prepared 1 solution at 0 °C and stirred for 2 h. To this yellow

diazonium salt solution, the 2-(N-ethylanilino)ethanol solution

(3.30 g, 0.02 mol) diluted with 10 mL of hydrochloric acid was

added slowly at 0 °C for 1 h and stirred at room temperature

for 20 h more. The red product mixture was poured into cold

water and sodium bicarbonate was added slowly to neutralize

the solution. The fine orange precipitate was filtered and

recrystallized from acetic acid, to give C2 as orange crystals.

Yield: 4.20 g (71%). UV (CH₃CN): λ_max459 nm. DSC (5 °C/

Eth yl-2-cya n o-2-(3-{(E)-2-[4-(d im eth yla m in o)p h en yl]-

1-eth en yl}-5,5-d im eth yl-2-cycloh exen yliden )a ceta te (C6).

Into a 100 mL flask were added 2.98 g (0.02 mol) of 4-N,N-

dimethyaminobenzaldehyde (12), 4.66 g (0.02 mol) of com-

pound 11, 40 mL of 1-propanol, and 10 drops of piperidine.

The solution was heated at 120 °C with stirring for 25 h, the

mixture was stored in a refrigerator overnight, and the crystals

of C6 were collected by filtration, washed with petroleum

ether, and dried. Yield: 2.9 g (40%). UV (CH₃CN): λ_max470

nm. DSC (5 °C/min): mp 189.5 °C, onset 184.7 °C. ¹H NMR

(CDCl₃): δ 7.41 (d, 2H, ArH), 6.97-6.77 (m, 4H, CHdCH and

ArH), 4.27 (q, 2H, CH₂), 3.04 (s, 3H, NCH₃), 2.98 (s, 2H, cyclic

CH₂), 2.40 (s, 2H, cyclic 2H), 1.37 (t, 3H, CH₂CH₃), and 1.04

(s, 6H, (CH₃)₂).

Syn th esis of NLO Nor bor n en e Mon om er s M1)M9. See

Supporting Information for these syntheses.

Syn th esis of NLO P oly(n or bor n en es). P r ep a r a tion of

P olym er iza tion Ca ta lyst Solu tion s. Preparation of catalyst

solution A for polymerization of HP , BP a through BP l, and

CP 1 through CP 9 was performed on a 2 mmol scale as

follows: In an oxygen- and moisture-free glovebox, 0.098 g (0.2

mmol) of the nickel catalyst and 10 mL of toluene were placed

in a 10 mL serum bottle to give a dark brown solution. The

bottle was tightly sealed. Preparation of catalyst solution B

was performed for 20-40 mmol scale polymerization of LCP 1

Macromolecules, Vol. 37, No. 14, 2004

Vinyl-Addition Poly(norbornene)s 5167

Ta ble 1. Resu lts of th e Tr a n sester ifica tion Stu d y Usin g

Ti or Sn Ca ta lysts

Ta ble 3. Resu lts of Blen d P olym er iza tion of

Hexyln or bor n en e in th e P r esen ce of 20 m ol % of a

Va r iety of Nitr ogen -Con ta in in g Com p ou n d s^a

reaction condition^a

13/DR1

ratio

cat. concn

(mol %)

time

(h)

convn^b

(%)

yield

(%)^c

no.

catalyst

T1

T2

T3

T4

T5

T6

T7

1/1

2/1

1/1

2/1

3/1

Ti

Sn

1

5

4

24

4

>50

>70

>80

>70

>90

>95

45

65

71

78

4

a

NB: 0.01 mol; toluene 50 mL; N₂; reflux; Ti, titanium

isopropoxide; Sn, dibutyltin oxide. Checked by LC. ^cPurified by

b

chromatography, followed by recrystallization.

Ta ble 2. P h ysica l P r op er ties a n d Exo/En d o Ra tios of

Nor bor n en e F u n ction a lized Mon om er s Used in Th is

Stu d y

guest

b

norbornene

monomer

mp^a

(°C)

bp

λ_max

exo^cendo^c

b

no.

HP

BP a

BP b

BP c

BP d

BP e

BP f

compound yield^b(%) M_w(×10³) M_n(×10³) M_w/M_n

MW

(°C/Torr) (nm)

(%)

98

0

1731

346

5.01

13

14

15

16

17

18

M1

M2

M3

M4

M5

M6

M7

M8

M9

166.21

124.18

285.38

313.39

299.36

178.31

434.49

414.50

78/10

103/20

175/0.4

39.9

21.0

21.5

20.5

28.7

40.7

36.0

38.9

20.0

60.1

79.0

78.5

79.5

71.3

59.3

64.0

61.1

80.0

d

a

b

c

d

e

f

94

90

98

97

0

472

971

1371

1046

77

131

349

208

6.11

7.41

3.93

5.04

116/20

478

448

432

493

458

472

435

420

472

76.8

94.1

a

Monomer: 2 mmol. Guest: 0.5 mmol. Ni catalyst concentra-

tion: 0.02 mmol. Solvent: toluene, 6 mL. Temperature: room

temperature. Time: 24 h. After one precipitation

418.49 118.9

467.61 121.2

532.68

b

58.4

Ta ble 4. Resu lts of Blen d P olym er iza tion Resu lts in th e

442.60 159.8

20.6

32.5

33.8

37.1

79.4

67.5

66.2

62.9

P r esen ce of 20 m ol % N,N-Dieth yla n ilin e Der iva tives^a

432.51

408.49

514.66

94.4

94.0

a

b

Peak points measured by DSC. Absorption maxima mea-

sured by UV-vis spectroscopy in acetonitrile. ^cCalculated values

from ¹H NMR spectra. The ratio could not be calculated due to

d

peak overlap.

through LCP 4 as follows: In an oxygen- and moisture-free

glovebox, 0.98 g (2.0 mmol) of the nickel catalyst and 50 mL

of toluene was placed in a 100 mL serum bottle and then sealed

tightly.

Hom op olym er iza tion . Into a dried 20 mL serum bottle

were added 0.357 g (2 mmol) of 5-hexyl-2-norbornene (or

5-hexyl norbornene) (18), 5 mL of anhydrous toluene, and a

magnetic stirring bar. The bottle was sealed, and the solution

was deaerated with a stream of nitrogen introduced and

removed with syringe needles. Then the catalyst solution A

(1 mL) was injected all at once with vigorous stirring. The

solution became viscous within 2 min and after 24 h at room

temperature the viscous polymer solution was diluted with

toluene and about 1 g of ion-exchange resin (Amberlite IRC

718, to remove the catalyst) was added and the solution stirred

for 5 h. After vacuum filtration, the filtrate was added

dropwise into acetone with vigorous stirring. The precipitated

white pulplike poly(hexylnorbornene) was collected by suction

filtration and dried overnight in an oven at 80 °C. Yield: 0.349

g (98%). The weight-average molecular weight (M_w) was

determined to be 1 731 000.

guest

compound (mmol)

concn

yield^b

(%)

M_w

M_n

b

no.

HP

BP g

BP h

BP i

BP j

BP k

BP l

(×10³) (×10³) M_w/M_n

98

90

87

92

0

1731

971

81

346

131

28

5.01

7.41

2.89

3.12

g

h

i

j

k

l

0.5

276

86

0

92

359

91

3.96

a

Blen d P olym er iza tion . A representative example of a

blend polymerization that was run to evaluate the compat-

ibility of functional groups with the Ni catalyst follows. Into a

dried serum bottle were placed 0.357 g (2.0 mmol) of 5-hexyl-

2-norbornene (18) and 0.0606 g (0.5 mmol) of N,N-dimethy-

laniline followed by 5 mL of anhydrous toluene and a magnetic

stirring bar. The bottle was capped, sealed, and deareated with

argon for 20 min. The catalyst solution A (1 mL) was injected

with a syringe into the polymerization bottle with vigorous

stirring. After 24 h at room temperature, about 1 g of ion-

exchange resin (Amberlite IRC 718) was added into the

Ni catalyst concentration: 0.02 mmol. Solvent: toluene, 6 mL.

b

Temperature: room temperature. Time: 24 h. After two pre-

cipitations.

polymerization bottle and stirred for 5 h. After filtration, the

polymer solution was added dropwise to acetone while stirring.

The precipitated white poly(norbornene) was collected by

filtration and dried overnight in oven at 80 °C. Yield: 0.335 g

(94%). The weight-average molecular weight (M_w) was deter-

mined to be 472 000. The other blend polymerizations (Bp a

through BP l in Table 3 and Table 4) were carried out using

5168 Park et al.

Macromolecules, Vol. 37, No. 14, 2004

Ta ble 5. NLO Cop olym er iza tion Resu lts of CP 1 fr om M1^a

no.

x/y

18 (mmol) M1 (mmol) toluene (mL) catalyst (mmol) yield^b(%) m/n^c(%/%) M_w(×10³) M_n(×10³) M_w/M_n

CP 1a

CP 1b

CP 1c

CP 1d

CP 1e

CP 1f

CP 1g

CP 1h

HP

80/20

100/0

80/20

70/30

60/40

1.6

2.0

1.6

1.4

1.2

1.0

0.4

0.0

0.4

0.6

0.8

1.0

8

6

4

2

6

0.02

0.01

0.02

0.03

0.04

0.02

70

76

70

37

74

76

65

59

98

76

65

36

19

99

125

107

53

128

125

95

47

56

49

25

50

56

38

43

346

56

44

32

33

2.09

2.22

2.18

2.16

2.58

2.22

2.50

2.18

5.01

2.22

1.90

1.73

2.12

94

100.0/0.0

83.3/16.7

76.5/23.5

69.9/30.1

64.3/35.7

1731

125

84

56

69

CP 1i

CP 1j

CP 1k

CP 1l^d50/50

a

b

Polymerization temperature: room temperature. Polymerization time: 24 h. After two precipitations. ^cMeasured by ¹H NMR

d

spectrometry. Polymerization time: 3 days.

Ta ble 6. Resu lts of Cop olym er iza tion of NLO

F u n ction a lized Mon om er s w ith Hexyln or bor n en e

(Mon om er : 2 m m ol Sca le)^a

the same procedure as described above except other additives

were substituted for the N,N-dimethylaniline.

Cop olym er iza tion . Several representative examples of

copolymerization follow in which more than one norbornene

monomer was used. All of the initial copolymerizations de-

scribed here were run using a total of 2 mmol of monomers.

LCP 1(8/2). Into a dried 20 mL serum bottle were placed

0.285 g (1.6 mmol) of 5-hexyl-2-norbornene (18), 0.174 g (0.4

mmol) of M1, 5 mL of anhydrous toluene, and a magnetic

stirring bar. This bottle was sealed and deareated for 20 min

using nitrogen. The catalyst solution A (1 mL) was injected

all at once into the polymerization bottle with vigorous stirring.

After 24 h at room temperature, the viscous polymer solution

was diluted with toluene, about 1 g of ion-exchange resin

(Amberlite IRC 718) was added to remove Ni catalyst, and the

mixture was stirred for 5 h. After vacuum filtration, the dark

red polymer solution was added dropwise into acetone while

stirring. The precipitated copolymer was dissolved again in

toluene and this polymer solution was precipitated once more

in acetone. The red powder was collected by filtration, washed

thoroughly with acetone until the red color in the acetone wash

could not be detected, and dried overnight in an oven at 80

°C. Yield: 0.41 g (73%). The weight-average molecular weight

(M_w) was determined to be 125 000.

no.

R

x/y

yield (%) M_w(×10³) M_n(×10³) M_w/M_n

CP 1(8/2)

CP 1(7/3)

CP 2(8/2)

CP 2(7/3)

CP 3(8/2)

CP 3(7/3)

CP 4(8/2)

CP 5

CP 6

CP 7

CP 8

CP 9

1

2

3

4

5

6

7

8

9

80/20

70/30

80/20

70/30

80/20

70/30

80/20

95/5

76

65

82

74

76

69

78

0

125

84

56

44

69

46

69

2.22

1.90

3.90

4.07

4.73

4.21

3.30

270

186

328

253

386

60

101

a

x + y ) 2 mmol, polymerization temperature: RT, polymer-

ization time: 24 h.

8.16 g (89%). The weight-average molecular weight (M_w) was

determined to be 123 000.

LCP 2(7/3). Into a dried 100 mL serum bottle were placed

2.496 g (14 mmol) of 2-hexyl-5-norbornene (18) and 2.511 g (6

mmol) of M3, 50 mL of anhydrous toluene, and a magnetic

stirring bar. This bottle was sealed and deareated for 30 min

with nitrogen. The catalyst solution B (5 mL) was injected all

at once into a polymerization bottle with vigorous stirring.

After polymerization at room temperature for 24 h, the viscous

polymer solution was diluted with toluene and about 10 g of

ion-exchange resin (Amberlite IRC 718) was added to remove

catalyst and the mixture was stirred for 5 h. After vacuum

filtration, the dark red polymer solution was added dropwise

to acetone while stirring. The precipitated red powder was

collected by filtration and dried overnight in oven at 80 °C.

Yield: 4.33 g (87%). The weight-average molecular weight (M_w)

was determined to be 185 000.

The other 2 mmol-scale copolymerizations (CP 1a through

CP 1l in Table 5 and CP 1(7/3) through CP 4(8/2) in Table 6)

were carried out using this same general procedure. After

initial screening on a 2 mmol scale some of the polymerizations

were scaled up to a preparative level of LCP (20-40 mmol

scale).

LCP 1(8/2). Into a dried 200 mL serum bottle were placed

5.706 g (32 mmol) of 5-hexyl-2-norbornene (18) and 3.476 g (8

mmol) of M1, 100 mL of anhydrous toluene, and a magnetic

stirring bar. This bottle was sealed and deareated for 30 min

with argon. The catalyst solution B (10 mL) was injected all

at once into a polymerization bottle with vigorous stirring.

After polymerization at room temperature for 24 h, the viscous

polymer solution was diluted with toluene and about 20 g of

ion-exchange resin (Amberlite IRC 718) was added to remove

catalyst and the mixture was stirred for 5 h. After vacuum

filtration, the dark red polymer solution was added dropwise

to acetone while stirring. The precipitated copolymer was

dissolved again in toluene, and this polymer solution was

precipitated again in acetone. The red powder was collected

by filtration and dried overnight in an oven at 80 °C. Yield:

Resu lts a n d Discu ssion

Syn th esis of Ch r om op h or es. The synthetic routes

of all the NLO chromophores involved in this study are

shown in Figure 2 either in part a or part b. These

chromophores are all alcohol or ester functionalized so

as to eventually attach to a complementary ester or

Macromolecules, Vol. 37, No. 14, 2004

Vinyl-Addition Poly(norbornene)s 5169

F igu r e 2. (a) Synthesis of the NLO chromophores that contain a link through the donor amine group. (b) Synthesis of the NLO

chromophores that contain a link through a site other than the donor amine group.

alcohol-functionalized norbornene by transesterification

chemistry. Chromophore C1 (Disperse Red 1, or DR1)

was commercially available while chromophores C2-

C6 were synthesized by standard diazonium coupling

or by Knoevenagel condensation. The NLO chro-

mophores in this study contain an azobenzene, stilbene,

or isophorone-derived π-conjugated bridge, a N,N-di-

methylaniline or a N,N-diethylaniline unit as an elec-

tron donor and a nitro, cyano, dicyanovinyl, or cyano-

estervinyl group as an electron acceptor. These types

chromophores are well-known, are well characterized,

and have been employed in numerous other nonlinear

optical polymers.^20,21

The starting material 3 was synthesized by a reaction

of N-methylaminoethanol with an excess of 4-fluoroben-

zaldehyde in the presence of a phase transfer agent at

90 °C. At higher temperature, byproducts increased

significantly, which were found to be due to aromatic

nucleophilic substitution of both amine and alcohol.

Compound 8 was synthesized by a condensation reaction

of 6 with 7 using potassium carbonate as a base, and

compound 11 was prepared by a condensation reaction

from isophorone.

Azobenzene C2 was prepared by a general diazonium

coupling procedure.²²Both C3 and C5 were synthesized

by Knoevenagel condensation using 4, the yields of

which were low despite use of excess 4 (which tends to

decarboxylate to 4-nitrotoluene). Chromophore C5 hav-

ing both a longer aliphatic group and different site of

attachment than C3 was synthesized so that the influ-

ence of a longer spacer group on poling could be

evaluated. The chemical structures of all these chro-

mophores were confirmed by ¹H NMR spectroscopy and

they all showed one sharp endothermic melting peak

except for C4 that exhibited two peaks due to two

crystal forms.

Syn th esis of F u n ction a lized NLO Mon om er s.

Details regarding the synthesis of the NLO monomers

(M1-M9) are given in the Supporting Information. The

chromophore-functionalized norbornene monomers were

synthesized by a transesterification reaction between

norbornene ethyl ester (13) with a hydroxy-substituted

chromophore (C1-C5) to give M1-M5 or between

norbornenemethanol (14) with the ester-substituted

chromophore C6 to give M6. These reactions schemes

are depicted in Figure 3 parts a and b. Since both

titanium isopropoxide and dibutyltin oxide are well-

known catalysts for transesterifications, as in the case

of the synthesis of poly(ethylene terephthalate), we

adopted these two catalysts for the synthesis of the

norbornene-functionalized monomers. To determine the

optimum transesterification conditions, the preparation

of M1 monomer by reaction of 13 with C1 (DR1) under

various reaction conditions was examined and the

results are summarized in Table 1. Both titanium

isopropoxide and dibutyltin oxide were useful for trans-

esterifications, but the tin catalyst usually gave a little

higher conversion (compare T4 with T6). Because

scrupulously anhydrous reaction conditions were not

used in our experiments the more moisture-sensitive

titanium isopropoxide was less effective. The dibutyltin

oxide was preferable because it is not as sensitive to

moisture or oxygen. As the 13/C1 ratio was increased

to 3/1 (T7), the yield increased. After a prolonged

exchange reaction time over 24 h, some byproducts due

to destruction of the chromophore began to accumulate.

5170 Park et al.

Macromolecules, Vol. 37, No. 14, 2004

F igu r e 3. (a) Synthesis of NLO norbornene monomers by transesterification via a link through the donor amine group. (b)

Synthesis of NLO norbornene monomers by transesterification via a link through a site other than the donor amine group. (c)

Synthesis of NLO norbornene monomers by transesterification and subsequent Knoevenagel condensation.

Macromolecules, Vol. 37, No. 14, 2004

Vinyl-Addition Poly(norbornene)s 5171

Using reaction conditions (T7) involving a feed ratio of

13/C1 ) 3/1, a catalyst concentration of 5 mol % and a

reflux temperature for 4 h, the M1 monomer could be

obtained in a 78% yield after purification by silica gel

chromatography. The excess of norbornene ester start-

ing compound could be easily removed during the

chromatography.

melting points of all solid monomers were determined

by DSC and all these samples showed one sharp

endothermic melting peak despite their isomeric com-

position.

Hom op olym er iza tion . We selected the norbornene

derivative 5-hexyl-2-norbornene (18) to make the ho-

mopolymer (HP ) and also as a comonomer, because the

n-hexyl functional group is inert to the catalyst and also

because it provides some extra solubility to any poly-

mers that contain it. First of all, we carried out the

homopolymerization of hexylnorbornene 18 (2 mmol)

with (η⁶-toluene)Ni(C₆F₅)₂(0.02 mmol, 1 mL of catalyst

solution A) in an inert atmosphere. After isolation, the

weight-average and number-average molecular weight

(M_wand M_n) for this homopolymer (HP ) as determined

by GPC were as high as 1 731 000 and 346 000, respec-

tively. The glass transition temperature could not be

detected by DSC, but instead the T_gwas determined to

be about 240 °C by dynamic mechanical analysis (DMA).

As a secondary structure modification, some chro-

mophores with a diethylamine donor group instead of

a dimethylamine donor were also prepared. There are

two reasons for this change: first, the ethyl-substituted

materials are slightly more nonlinear and, second, the

ethyl-substituted compounds are more thermally stable.²³

We attempted to synthesize 4-[(2-hydroxyethyl)(ethyl)-

amino]benzaldehyde by a reaction of 2-(ethylamino)-

ethanol with 4-fluorobenzaldehyde analogous to the

synthesis of 3, but the desired product was obtained in

poor yield due to the steric effect (ethyl vs methyl) in

the aromatic nucleophilic substitution reaction. There-

fore, the monomers M7 and M8 containing the N-ethyl

group found in 16 were synthesized instead by a route

involving first transesterification between 13 and 2 to

give 15 followed by post-formylation reaction to give 16

and finally the Knoevenagel reaction with 4 and 9

producing M7 and M8 respectively (Figure 3c). In the

Vilsmeier-Haack formylation reaction of 15, it was

fortunate that the norbornene ester acted as a good

protecting group.

Blen d P olym er iza tion . To determine the tolerance

of the nickel catalyst for various functional groups, we

attempted the homopolymerization of hexylnorbornene

monomer (18) in the presence of a variety of nitrogen-

containing guest compounds since most NLO chro-

mophores consist of nitrogen atom(s) at differing levels

of oxidation in either or both of the electron-donor and

electron-acceptor functional groups. The concentration

of each simple functionalized guest (see Table 3, a -f)

was 20 mol % relative to hexylnorbornene. This func-

tional tolerance test of (η⁶-toluene)Ni(C₆F₅)₂is intended

to provide some guidance in defining and designing

norbornene functionalized NLO monomers that are

compatible with the nickel system. The M_w, M_n, and

polydispersities (M_w/M_n) of the poly(hexylnorbornene)s

obtained in the presence of these additives are also

provided in Table 3.

All the chemical structures of the norbornene con-

taining monomers could be fully characterized by ¹H

NMR spectroscopy. The norbornene ethyl ester 13,

synthesized by Diels-Alder reaction, contained exo and

endo isomers, so of course the M1 monomer also

contains two isomers. Figure 4a shows the ¹H NMR

spectra of the isomer mixture (the exo, and endo isomers

of M1 monomer, respectively). Because the exo isomer

moved slightly faster than endo isomer on silica gel

chromatography (toluene), we could separate the two

red spots on preparative TLC plates for NMR analysis.

All the hydrogens in structure M1 were easily assigned

As a first obvious result it is clear that the polymer-

izations attempted in the presence of aniline (a ) or

benzonitrile (f) were unsuccessful. Both the primary

amine and cyano groups stop the polymerization, The

influence of the cyano group was particularly trouble-

some as it is a common structure acceptor group feature

of many (and especially some of the most nonlinear) of

NLO chromophores.²⁵It is entirely possible that the

cyano and amino groups render the catalyst inactive by

competing with the olefin for an open coordination site

on the metal. Stable complexes of the Ni(C₅F₆)₂frag-

ment and its palladium analogue coordinated with

benzonitrile and benzylamine have been reported.²⁶The

result with aniline itself is not so troublesome because

most NLO chromophores have fully substituted amines

as donors and the catalyst tolerated these amines. More

specifically, the polymerization in the presence of N,N-

disubstituted aniline guests (b , c, and d ) yielded

polymers at comparable yields but with lower M_w’s than

a blank polymerization without guest compound. Con-

sidering the highest M_wof BP d among BP b-BP d , we

note that the most-hindered aniline moiety, such as d ,

was preferable in nickel-initiated polymerization. These

results implied that the nickel compound easily formed

a complex with a primary amine in preference to

forming a reactive complex with the alkene functionality

of the monomer. Triphenylamine does not appear to

affect the polymerization very much since the M_wof the

homopolymer is similar to the polymerization without

a guest compound. This may be due to a combination

of the steric hindrance and/or the reduced basicity of

1

in the H NMR spectrum. The spectra of the exo and

endo isomers differ most around 6 ppm corresponding

CHdCH in the norbornene unit. The hydrogens in the

1- and 2-positions of the exo-norbornene (Figure 4a) had

two doublets of doublets due to coupling with the

adjacent two hydrogens. The hydrogens on positions 1

and 2 in the endo-norbornene unit had also two doublet-

doublets, but with a larger coupling constant than those

of the exo isomer.²⁴Figure 4b shows the NMR spectra

of the two isomers expanded around 6 ppm.

All the monomers (M1-M9) synthesized in this study

were a mixture of exo and endo isomers. This composi-

tion of the norbornene ethyl ester starting material (13)

is 39.9% exo and 60.1% endo by NMR. From the integral

of the isomer mixtures in Figure 4b the isomer content

was calculated to be 40.7% exo and 59.3% endo. In Table

2 the exo/endo ratio of these norbornene monomers are

tabulated as well as melting points, boiling points, and

maximum absorption wavelengths (λ_max). The isomer

ratio of M5 could not be calculated because the CHd

CH peaks overlapped with the aromatic peaks, but it

can be assumed that the ratio is not much different than

in their precursors.

Except for M9 all of the NLO monomers were ob-

tained as crystals or solids although they were mixtures

of exo/endo isomers. In some cases, such as for M4 and

M5, the monomers took a long time to solidify. The

5172 Park et al.

Macromolecules, Vol. 37, No. 14, 2004

F igu r e 4. (a) ¹H NMR (300 MHz) spectra of the M1 monomer stereoisomer mixture and the individual endo and exo components.

1

(b) Expanded H NMR (300 MHz) spectra of the olefinic region of the M1 monomer mixture and of the individual pure stereoisomers.

this amine that results in weaker coordination of the

amine to the nickel center. Considering the higher

molecular weight of BP c (N,N-diethyl) relative to BP b

(N,N-dimethyl) it appears that the steric effect must

again be a more critical influence in determining mo-

lecular weight than the basicity of aniline.

Examination of Table 3 also shows that the nitro

group is tolerated by (η⁶-toluene)Ni(C₆F₅)₂although once

Macromolecules, Vol. 37, No. 14, 2004

Vinyl-Addition Poly(norbornene)s 5173

again the molecular weight of BPe was lower than that

prepared in the absence of nitrobenzene, i.e., HP .

Among these tolerance tests, the successful polymeriza-

tions in the presence of nitro, N,N-dimethyl, N,N-

diethyl, and N,N-diphenyl units were very encouraging

because many NLO chromophores contain these same

structure units. On the basis of this preliminary exami-

nation, we deduce that the nickel catalyst tolerance

order (related to the molecular weight of polymer

produced) involving the following nitrogen-containing

functional groups is as follows: -NPh₂> NEt₂> NMe₂

> NO₂. NH₂∼ CN.

of Ni catalyst at room temperature for 24 h. In all cases,

the gel permeation chromatograms were unimodal

consistent with the formation of a copolymer.

With these same reaction conditions kept constant the

level of NLO chromophore incorporation into the co-

polymer (compare HP -CP 1l in Table 5) was examined,

i.e., does the composition of the copolymer reflect the

original feed ratio of the monomers? It was found that

as the chromophore-containing monomer M1 portion

increased both the molecular weights and the yields of

the copolymers steeply decreased. At a feed ratio of x/y

) 80/20 (mol %/mol %) the M_wdecreased by an order of

magnitude compared to the homopolymer M_w(HP ).

However, a copolymer (CP 1k ) polymerized at a feed

ratio as high as 50/50 (mol %/mol %) still had an M_wof

70 000. The DR1 content in copolymers was determined

As a next functional group tolerance test for the nickel

catalyst blend polymerizations in the presence of 20 mol

% of various N,N-diethyl-substituted NLO chromophores

were carried out. These guest compounds are all di-

ethylanilines that contain additional acceptor functional

groups such as nitro, aldehyde, cyano, dicyano, and

ketone in the para position (Table 4). J ust as in the case

of the first blend polymerization study (Table 3), all

polymerizations attempted on cyano group containing

monomers (BP j and BP k ) failed. On the basis of the

results already described (Table 3), we expected that

an electron-withdrawing nitro group would decrease the

basicity of aniline so that BP i might be obtained with

a higher M_wthan BP g. However, the outcome revealed

that the N,N-diethyl group and nitro group give a still

lower M_wthan BP g and this indicates again that the

relative basicity of the aniline does not appear to be a

significant influence. Nevertheless, the polymerization

BP i containing the nitro-substituted aniline gave a poly-

(norbornene) of M_w276 000 that certainly is high

enough for creation of nonlinear optical thin films and

therefore a very promising result. It was also found that

the aldehyde functional group lowered the M_wof the

polymer much more than a ketone functional group (at

least relative to the type of ketone found in indanedi-

one). On the basis of this preliminary examination, we

deduce that the nickel catalyst tolerance order (related

to the molecular weight of polymer produced) for the

following functional groups is: ArCdO > NO₂> CHO

. CN ∼ CdC(CN)₂. Although the ketone functional

group was preferable to the nitro group for obtaining a

high M_wof poly(norbornene), the latter group appears

more attractive considering it is a better electron

acceptor for higher NLO susceptibility. Thus far, the

nitro group appears to be the most practical acceptor

for NLO chromophores involved in polymerizations

using this nickel-initiated process.

1

by 500 MHz H NMR spectrometry by normalizing the

aromatic resonance at 8.3 ppm (2H) normalized to 1.00

(see Supporting Information). Between a feed ratio of

m/n ) 80/20-50/50 (mol %/mol %) the n value was in

the range of 17-36 mol % and so the actual loading of

the NLO chromophore was somewhat lower presumably

due to the lower polymerization reactivity of an NLO

monomer like M1 relative to 18. Considering the higher

loading of DR1 chromophore and the film forming

capabilities, CP 1i or CP 1j may be good candidates for

practical production among these copolymers. It is also

a promising sign that the two nitrogen atoms in the azo

linkage (-NdN-) did not critically affect the catalyst

activity.

Figure 5 shows all the norbornene functionalized

monomer structures synthesized in this study and the

specific results of copolymerization with hexylnor-

bornene are summarized in Table 6. All the polymeriza-

tions except CP 5-CP 9 were carried out on a 2 mmol

monomer scale (13/M1) of 80/20 (mol %/mol %) and 70/

30 (mol %/mol %) feed ratio, respectively. Because of

the nickel complex’s incompatibility with the cyano

group, the feed ratios of polymerization of CP 5-CP 9

were decreased to only 95/5 (mol %/mol %). Despite this

considerable reduction in concentration of cyano-

containing monomers all the polymerization attempts

of CP 5-CP 9 were unsuccessful indicative of the potent

inhibitory effect of the cyano group. The stilbene-based

NLO monomers were all successfully copolymerized

with 18 to give higher molecular weight NLO polymers

(CP 2-CP 4) than the comparable azobenzene-based

NLO polymer CP 1 which may be due to a weak

inhibitory influence on catalysis by the nitrogen con-

taining π-bridge (-NdN-) group. The higher M_wof

CP 3 compared with CP 2 was due to the steric demands

of the N,N-diethyl group vs the N,N-dimethyl group

with smaller alkyl substitution as was demonstrated in

the tolerance test. The highest M_wvalue of CP 4 may

be reflected by the bulkier structure of CP 4, because

the molecular weights were measure by size exclusion

chromatography.

Cop olym er iza tion . With the knowledge that the

nickel catalyst tolerated both tertiary amine and nitro

functional groups, we examined the conditions for

copolymerization of hexylnorbornene with the nor-

bornene functionalized DR1-based monomer (M1). The

feed ratio of two monomers was fixed at 80/20 (mol

%/mol %) and the influence of overall monomer concen-

tration on M_wwas examined (compare CP 1a -CP 1d ,

Table 5). The polymerization at 2 mmol monomer in 6

mL of solvent produced the highest M_wof 125 000. More

concentrated and more diluted systems produced lower

M_w. The optimum catalyst concentration was found to

be in the range of 0.5-1.0 mol % (compare CP 1e-

CP 1h ). When the catalyst concentration was increased

above 1 mol % the M_wof copolymer was not increased

significantly and so 1 mol % catalyst was sufficient. The

standard polymerization conditions for 2 mmol of mono-

mer was found to be 6 mL of toluene, 0.01-0.02 mmol

On the basis of the successful small-scale (2 mmol)

polymerization results larger scale polymerizations for

the preparation of sufficient material for electrooptical

evaluations were run and Table 7 shows the polymer-

ization results (LCP 1-LCP 4). The copolymers were

obtained in a 65-75% yield and with M_w102 000-

278 000 as red-orange powders. The molecular weight

values reproduced those of the small-scale polymeriza-

tion quite well. The 500 MHz ¹H NMR spectra of

LCP 1-LCP 4 are given in the Supporting Information.

5174 Park et al.

Macromolecules, Vol. 37, No. 14, 2004

F igu r e 5. Copolymerization of CP R(x/y). Here the acyl carbon is the attachment site of group R.

The aromatic peaks derived from the chromophores

appear between 6.0 and 8.3 ppm. Broad aliphatic peaks

from the spacers and the norbornane backbone appear

between about 4 and 0.8 ppm. No peaks due to unre-

acted norbornene monomer are present in the 6 ppm

region. From these five different NLO copolymers one

should be able to investigate the difference of NLO

properties between azobenzene and stilbene unit, chro-

mophore content, N-methyl and N-ethyl unit, and short

spacer vs long spacer units. All the LCP copolymers

synthesized were fairly soluble in toluene, chloroform,

dichloroethane, 1,1,2,2-tetrachloroethane (TCE), and

tetrahydrofuran, but they were poorly soluble or not

soluble in acetone, methanol, N,N-dimethylformamide,

and N-methylpyrrolidone. We tried to determine the

glass transition temperatures of NLO chromophore

substituted poly(norbornene)s by DSC (between ambient

and 300 °C), but just as in the case of the less

substituted parent systems, we were unable to observe

any glass transitions.

samples we have to consider the relative polymer film

thickness, the chromophore volume density (N), and the

nonlinear activity of each chromophore. The second

harmonic signal is proportional to the square of these

parameters, and in this experiment, all these param-

eters are carefully considered. A relative measurement

of N can be obtained from absorbance scans for both

films as depicted in Figure 6. At peak absorption for 1

µm thick films the optical density (OD) of DR1-nor-

bornene is 2.52 and for DR1-MMA it is 2.41. These

absorbance values match the estimated weight or

volume percentage of active chromophore in LCP 1 at

25.9% and DR1-MMA polymer at 23.2% nicely. We will

assume that the chromophore nonlinear activity is

comparable in the two different polymers. The amino-

nitroazobenzenes are nonlinear dyes with positive sol-

vatochromism, and in 10% DR1-MMA the λ_maxis at

about 475 nm while the 20% LCP1 has a λ_maxat about

450 nm. The hydrocarbon rich norbornene environment

is probably less polar than the ester rich acrylate

environment and the chromophore is probably some-

what more nonlinear in the latter polar environment.

Before looking at the details of the poling experiments

a comparison of the chromophore loading in these

polymers is necessary. On the basis of monomer feed

the LCP 1 polymer has a 20 mol % chromophore content

(determined experimentally to be about 18.8%, very

Non lin ea r Op tica l Ch a r a cter iza tion . The effective

second-order susceptibility ø⁽²⁾of polymer LCP 1(8/2)

(from here on referred to simply as LCP 1) was mea-

sured in comparison with DR1-MMA copolymer (10 mol

% chromophore, 90 mol % MMA) using second harmonic

generation. To compare the magnitude of the second

harmonic signals between the two different polymer

Macromolecules, Vol. 37, No. 14, 2004

Vinyl-Addition Poly(norbornene)s 5175

Ta ble 7. NLO Cop olym er iza tion Sca le-u p Resu lts^a

no.

R

x/y (mol %/mol %)

yield (%)

m/n (mol %/mol %)

M_w(×10³)

M_n(×10³)

M_w/M_n

LCP 1(8/2)

LCP 1(7/3)

LCP 2(7/3)

LCP 3(7/3)

LCP 4(7/3)

1

2

3

4

80/20

70/30

76

65

74

69

73

81.2/18.8

71.6/28.4

69.7/30.3

69.1/30.9

68.9/31.1

123

102

185

191

278

43

38

47

48

75

2.90

2.66

3.94

4.01

3.70

a

b

Polymerization temperature: RT, polymerization time: 24 h. Measured by ¹H NMR spectra.

involved in the definition applies to both systems.

Whenever NLO polymers are compared, especially

across different polymer classes, this kind of chro-

mophore loading analysis is required. Different polymer

films of equal thickness containing identical chro-

mophores should generate second harmonic signals of

equal magnitude if the chromophore loadings are com-

parable, the extent of chromophore orientation is com-

parable and the microscopic properties (the µâ products)

are comparable. All these contributions have been

considered in the following four sets of poling experi-

ments that compare DR1-MMA and LCP1.

The first set of poling experiments was performed in

a poling station without in situ SHG measurements. A

total of four films, two DR1-MMA (#1, #2) and two

LCP 1 (#3, #4), with thicknesses well below their coher-

ence length were prepared and corona poled. The

coherence length is l_c) (λ/2)(1/(n^2ω- n^ω)), where λ is

the fundamental laser wavelength and n^ωand n^2ωare

the refractive indices at the fundamental and second

harmonic wavelengths, respectively of DR1-MMA and

LCP 1 at 1057 nm are 3.95 and 4.26 µm, respectively.

For each film the P^2ωwas obtained at two different

points during the overall poling process: the first

determination of P^2ωis done after cooling to room

temperature but prior to removal of surface charge while

the second determination of P^2ωis done after elimination

of surface charge by immersion of the film in water for

5-10 min. The first set of poling experiments was

performed in a poling station without in situ SHG

measurements. The poling experiments the conditions

chosen for LCP 1 were a temperature in the 100-120

°C range, voltage ∼6 kV and a current in the 3-4 µΑ

range. The ratios of susceptibilities among these four

films was calculated from the eq 2 are summarized in

Table 8a, while Table 8b shows all the parameters used

that are used in the calculation.

F igu r e 6. UV-vis spectrum of thin films of DR1-MMA and

LCP 1.

close to theoretical, by NMR, Table 7) while the DR1-

MMA polymer has a 10 mol % monomer feed. Although

the mol % of the chromophores differ by a factor of

2 the weight or volume ratio is much closer to one. To

make this comparison the LCP 1 polymer was dis-

sected into three parts, the hexylnorbornene comonomer

(C₁₃H₂₂, MW 137), the norbornene acid part of the ester-

containing monomer (C₈H₉O₂, MW 137), and, finally,

the “active” part 4-diethylamino-4′-nitroazobenzene

(C₁₆H₁₈N₄, MW 297). The weight or volume percentage

of active chromophore in LCP 1 is thus 25.9% given by

the ratio (2 × 297)/(8 × 178 + 2 × 137 + 2 × 297). A

comparable analysis for the DR1-MMA polymer was

done, and here the polymer containing 10 mol %

chromophore monomer was determined to be 23.2%

weight or volume active chromophore from the ratio (1

× 297)/(9 × 100 + 1 × 85 + 1 × 297). So, on volume or

weight percentage basis, the two polymers that will be

compared here have similar loadings of the same

chromophore on a weight or volume basis. The definition

of the “active” and “inactive” components of these

polymers is somewhat arbitrary, but whatever bias is

In Table 8a the ratio of the effective susceptibilities

of the two DR1-MMA films (#1 and #2) before and after

removal of surface charge (leading to removal of the field

resulting from the surface charge) are both close to 1,

indicating that the experimental premise that we can

5176 Park et al.

Macromolecules, Vol. 37, No. 14, 2004

Ta ble 8. Ra tio of Su scep tibilities a n d P a r a m eter s Used ,

Wh er e th e Su bin d ex In d ica tes th e F ilm Nu m ber

Ratio of Susceptibilities

immediately

after poling

after neutralization

of surface charge

(2)

ø₁⁽²⁾/ø₂

ø₁⁽²⁾/ø₃

ø₁⁽²⁾/ø₄

ø₂⁽²⁾/ø₃

ø₂⁽²⁾/ø₄

1.06

1.17

1.19

1.12

1.13

1.1

2.32

2.01

2.13

1.68

Parameters Used for Calculation of Susceptibility Ratios

film no.

EOP

L (nm)

P^ω(mW)

P

^2ω(au)

1

2

3

4

DR1-MMA

LCP1

430

500

455

415

590

610

590

13.9

17

2.9

3.2

LCP1

pole DR1-MMA in a repeatable manner is correct.

However, and in contrast, the effective second-order

nonlinear optical susceptibility ratios between the DR1-

MMA films (#1 and #2) and LCP 1 films (#3 and #4) are

all around 1 before water treatment but closer to 2 after

water treatment. In DR1-MMA only a small reduction

in the effective second-order susceptibility occurs upon

removal of surface charge (on the order of 10%) while

in the LCP 1 films this reduction in the effective second

order susceptibility is much larger (on the order of 50%)

upon removal of the surface charge. A change (reduc-

tion) in the effective susceptibility can occur upon

removal of surface charge due to (at least) two contribu-

tions: first, when the surface charge is removed then

the EFISH contribution is eliminated and, second, if the

nonlinear molecular dipoles are not firmly held by the

polymer matrix then dipoles may relax to a less ordered

orientation. Ideally, when the surface charges are

removed the polymer matrix should hold the dipole

orientation and only the EFISH contribution would be

eliminated. The results of this experiment indicates that

in comparison with DR1-MMA, in LCP 1 this “chro-

mophore trapping in the polymer matrix” is not nearly

as effective as it should be, resulting in additional

dipolar relaxation.

Because of these very unusual chromophore relax-

ation results with LCP 1 a second set of poling experi-

ments were repeated using an in situ device so that the

optimum poling temperature could be better observed

by direct determination of the SHG signal. These in situ

experiments are intended only to identify the most

advantageous conditions to obtain maximum P^2ωbut not

to determine absolute or even relative ø⁽²⁾values. The

suspicion here was that in the first set of measurements

that the poling for LCP 1 was not performed in the

appropriate glass transition region so this second in situ

SHG experiment as a function of temperature was

performed. For the in situ poling studies a 2.4 µm thick

film of LCP 1 polymer was spin-coated onto the backside

of 0.8 mm thick ITO-coated substrates. After spinning,

the film was baked for >12 h in a vacuum oven at 100

°C, and after cooling, the film was mounted in the in

situ device. The corona discharge was performed with

a fixed voltage of 5 kV. The film was gradually heated

while the SHG and poling current were measured. As

can be seen in Figure 7 for a fixed poling voltage the

SH signal grows steadily as temperature increases from

ambient to about 140 °C, then a broad maximum of

about 14 au is obtained around 140-160 °C and then

finally the SH signal diminishes above 160 ^°C. Likewise,

F igu r e 7. SHG as a function of temperature for the LCP 1

nonlinear optical polymer.

F igu r e 8. Poling current as a function of temperature for the

LCP 1 nonlinear optical polymer.

in Figure 8 it can be seen that the current through this

same film gradually increases as the temperature is

raised between ambient and 160 °C and then rises

abruptly above 160 °C. In other polymers studied

previously this simultaneous drop in SHG and rise in

poling current at a common temperature is the signa-

ture of a glass transition. As the glass transition is

approached the chromophore mobility and the ability

to align in response to the field both increase. However,

the ionic mobility rises simultaneously and the in-

creased current ultimately compromises the magnitude

of the electric field across the film. For this reason, and

because the poling efficiency always contains a compo-

nent that is inversely proportional to kT, there is no

advantage in raising the poling temperature still fur-

ther.

From this second set of in situ experiments on LCP 1

just described it appears that the glass transition of this

poly(norbornene) might be higher than the poling tem-

perature used in the first set of experiments. So, after

the in situ poling studies we hoped to verify that at a

160 °C poling temperature we might now be able to

better pole LCP 1 and on cooling to more effectively

“lock” the oriented chromophore in the matrix. For the

follow up third set of experiments an approximately 1:1

solution (0.2 µm filtered) of LCP 1 in 1,1,1,2-tetrachlo-

roethane was spin-coated on a 1 mm thick sodalime

Macromolecules, Vol. 37, No. 14, 2004

Vinyl-Addition Poly(norbornene)s 5177

glass substrate at 4000 rpm. After baking in a vacuum

for approximately 12 h this new film 5 had a thickness

of 455 nm that is well below its coherence length (4.26

µm). This film was poled for 30 min in the poling station

at the “optimal” temperature of 160 °C using 5.9 kV

poling voltage and 2.6 ( 0.2 µA poling current. The

poling voltage was determined using the in situ cell by

observing the growth of second harmonic signal as the

poling voltage is raised to a point were the growth stops.

J ust as in the case of films 1-4 after the optimum poling

parameters have been determined in the in situ cell the

poling station is used because it can provide poling

parameters that can be controlled more accurately. After

exposure to elevated temperature and high voltage this

film was allowed to gradually cool over approximately

2 h while the high voltage was sustained. Immediately

after poling (and prior to immersion of the film in water)

a second harmonic signal of 11.8 au was measured.

However, after 5 min of immersion in DI water, a second

harmonic signal of 3.2 au was measured for the same

film at the same fundamental laser power and PMT

voltage. The ratio of effective second-order nonlinear

optical susceptibility is therefore 0.51, which is consis-

tent with the first set of poling experiments where this

ratio, calculated from Table 8a, is 0.57 ( 0.08. These

results indicate that “locking” of the chromophore in the

LCP 1 polymer backbone matrix was not achieved

even after poling at the higher poling temperatures of

160 °C.

F igu r e 9. SHG obtained from previously unpoled DR1-MMA

and LCP 1 films under the influence of a 5 kV corona-poling

field at room temperature.

the host does play a role in this regime, it signal

strength is dominated by the charge build-up on the

film. However, after the initial large growth of second

harmonic signal the LCP1 film the signal continues to

increase and during the 3500 s duration of the experi-

ment it never reaches a stable level, while the DR1-

MMA signal plateaus after the initial large growth. This

implies that in the case of LCP 1 that either the electric

field continues to increase and/or the chromophore

reorientation is taking place in LCP 1 at room temper-

ature (and indicative of chromophore mobility even well

below the apparent T_g). While there certainly are

differences in the dielectric properties of DR1-MMA and

LCP 1, one does not expect them to be so different as to

explain the difference in the electric field build-up.

While it is known that a chromophore can be mobile

well below the glass-transition temperature,²⁷it is of

interest here that the same chromophore is much more

mobile in the poly(norbornene) host than in the meth-

acrylate host, indicative of the existence of more free-

volume in the poly(norbornene). This experiment cor-

roborates the idea that chromophore polar order

relaxation is the source of the large orientation decay

seen in freshly poled DR1-norbornene samples after

immersing them in DI water. The greater chromophore

mobility in the norbornene system leads to loss of

chromophore orientation after the residual charges

introduced by the poling procedure are removed.

From these experiments, we observed that the poly-

(norbornene) matrix in LCP 1 behaves quite differently

than the methacrylate matrix in DR1-PMMA and, in

fact, quite differently from any other polymers in our

experience. The reasons for this behavior is not yet fully

understood; however, we conjecture that at the higher

temperature that the void space is not filled in the

LCP 1 polymer, and this makes it easier for the reori-

entation of the nonlinear chromophore in the absence

of an electric field upon cooling and removal of surface

charges.

Finally, an additional fourth set of experiments were

devised, in this case to evaluate the relative room

temperature mobility of the DR1 chromophore when it

is incorporated in the PMMA and poly(norbornene)

systems. In this fourth experiment, previously unpoled

(isotropic) films of both DR1-MMA and LCP1 are

maintained at room temperature under the influence

of a corona discharge electric field. Any SH signal that

results will be due only to EFISH if the molecules

cannot reorient, but if there is some freedom for molec-

ular rotation at room temperature (and well below T_g

of either polymer), then an electrooptic component will

also contribute to SHG. Here, a 3.9 µm film of LCP-1

and a 4.0 µm film of DR1-MMA were spin-coated on the

uncoated surface of a ITO-coated glass substrates. The

film thickness for the films is almost identical. Both

films were baked under vacuum at 100 °C for a period

of at least 24 h. The films were mounted the in situ

poling cell, and at room temperature a 5 kV voltage was

applied producing a current of less the 0.2 µA. Prior to

the conduction of an experimental run it was verified

that the electrical poling conditions were identical for

both films. The SH signal was sampled every 15 s for a

period of 3600 s. To avoid any possible photodegradation

damage to the chromophore (this is only a precautionary

measure as we do not have any evidence of degradation)

due to the long duration of the experiment, a laser power

lower than the first three sets of experiments was used

and the laser was blocked by a shutter that periodically

opened only for 2 s (enough time for the lock-in to settle).

The results of this fourth experiment are very clear

from examination of Figure 9. During the first 500 s of

poling the signals from DR1-MMA and LCP1 are

comparable. The second harmonic signal does not

increase instantaneously because the corona-discharge

poling configuration has a large time constant due to

large product of the effective resistance and effective

capacitance. While the mobility of the chromophore in

Con clu sion s

In this study, we have adopted the addition polym-

erization of norbornenes as a method for the preparation

of electroactive polymers, in particular polymers de-

signed as electrooptical materials. Synthetic pathways

have been developed that permit the introduction of the

5178 Park et al.

Macromolecules, Vol. 37, No. 14, 2004

(8) (a) Sartori, G.; Ciampelli, F.; Cameli, N. Chim. Ind. (Ital.)

1963, 45, 1479. (b) McKeon, J . E.; Starcher, P. S. US Patent

3,330,815, 1967.

reactive functionality, a norbornene ring, into a wide

range of second-order nonlinear optical chromophores.

The polymerization of these norbornene-functionalized

monomers with a nickel catalyst has been examined

and with the help of guest-host model systems the

tolerance of the nickel catalyst to typical chromo-

phore acceptor and donor functional groups has been

determined. In the case of a hexylnorbornene and

DR1 containing material, the synthesis was success-

fully scaled up to the multigram level to give an amino-

nitro-substititued azobenzene material LCP 1 with good

film forming and other good ancillary properties for

NLO application. Preliminary measurements of SHG

in this poly(norbornene) material indicate that it can

be poled just like a conventional system such as DR1-

MMA but that the poled order stability appears to suffer

from enhanced relaxation. The polar order relaxation

in LCP 1 was studied in some detail including a room-

temperature poling experiment which may be of

general utility for evaluation of poled order stability in

polymers. The relaxation phenomena found in LCP 1

needs to be examined in other related norbornene

systems to see if it is very general. If so, subsequent

studies to mitigate (or possibly exploit) this relaxation

may be required.

(9) Much of the work of Kaminsky, Sen, Risse, Heitz, Novak,

Arndt, and Goodall has been reviewed recently. (a) J aniak,

C.; Lassahn, P. G. J . Mol. Catal. A: Chem. 2001, 166, 193.

(b) J aniak, C.; Lassahn, P. G. Macromol. Rapid Commun.

2001, 22, 479-492. (c) Goodall, B. L. In Late Transition Metal

Polymerization Catalysis; Rieger, B., Saunders Baugh, L.,

Kacker, S., Striegler, S., Eds.; Wiley-VCH: Weinheim, Ger-

many, 2003.

(10) (a) Arndt, M.; Gosmann, M. Polym. Bull. (Berlin) 1998, 41,

433. (b) Arndt, M.; Kaminsky, W. Macromol. Symp. 1995,

97, 225.

(11) (a) Goodall, B. L.; Benedikt, G. M.; McIntosh, L. H., III;

Barnes, D. A. US Patent 5,468,819, 1995. (b) Goodall, B. L.;

Benedikt, G. M.; McIntosh, L. H., III; Barnes, D. A.; Rhodes,

L. F. US Patent 5,569,730, 1996. (c) Goodall, B. L.; Risse,

W.; Mathew, J . P. US Patent 5,705,503, 1996. (d) Goodall,

B. L.; Benedikt, G. M.; McIntosh, L. H., III; Barnes, D. A.;

Rhodes, L. F. Proc. Am. Chem. Soc. Div. Polym. Mater.: Sci.

Eng. 1997, 76, 56. (e) Goodall, B. L.; Benedikt, G. M.;

McIntosh, L. H., III; Barnes, D. A.; Rhodes, L. F. Polym.

Prepr. (Am. Chem. Soc., Div. Polym. Chem.) 1998, 39 (1), 216.

(12) Ivin, K. J .; Mol, J . C. Olefin Metathesis and Metathesis

Polymerization; Academic Press: San Diego, CA, 1997; p 407.

(13) Barnes, D. A.; Benedikt, G. M.; Goodall, B. L.; Hullihen, K.;

J ayaraman, S.; McDougall, W. C.: McIntosh, L. H., III;

Rhodes, L. F.; Shick, R. A. Proceedings of MetCon 98

(Worldwide Metallocene Conference); Catalyst Consultants

Inc.: Houston, TX, 1998.

(14) (a) Kim, J . M.; Shin, H. Y.; Park, K. H.; Kim, T. H.; J u, S. Y.;

Han, D. K.; Ahn, K. D. Macromolecules 2001, 34, 4291. (b)

Kang, S. H.; Shin, H.-D.; Oh, C.-H.; Choi, D. H.; Park, K. H.

Bull. Korean Chem. Soc. 2002, 23, 957.

(15) Rhodes, L. F.; Bell, A.; J ayaraman, S.; Lipian, J .-H.; Goodall,

B. L.; Shick, R. A. US Patent 6,232,417, 2001.

Ack n ow led gm en t. The investigations at Promerus

LLC (formerly BFGoodrich Electronic Materials) and

Kent State University were supported by an Advanced

Technology Program grant from the National Institute

of Standards and Technology (ATP Project 98-02-0023).

This work at UC Davis was supported in part by the

MRSEC Program of the National Science Foundation

under Award DMR-9808677 for CPIMA II.

(16) Pretre, Ph.; Wu, L. M.; Hill, R. A.; Knoesen, A. Opt. Soc. Am.

B 1998, 15, 379.

(17) Singer, K. D.; Lalama, S. L.; Sohn, J . E.; Small, R. D.

Electrooptic Organic Materials, Chapter II-8. In Nonlinear

Optical Properties of Organic Molecules and Crystals; Chem-

la, D. S., Zyss, J ., Eds.; Academic Press: 1987; Volume 1.

(18) Knoesen, A. Chapter 9: Photoisomerization in Polymer Films

in the Presence of Electrostatic and Optical Fields. In

Photoreactive Organic Thin Films; Sekkat, Z., Knoll, W., Eds.;

Academic Press: Amsterdam, 2002.

Su p p or tin g In for m a tion Ava ila ble: Text giving the

synthesis of norbornene monomers M1-M9 and figures show-

ing NMR spectra of norbornene copolymers. This material is

available free of charge via the Internet at http://pubs.acs.org.

(19) Lupo, D.; Ringsdorf, H.; Schuster, A.; Seitz, M. J . Am. Chem.

Soc. 1994, 116, 10498.

(20) Characterization Techniques and Tabulations for Organic

Nonlinear Optical Materials; Kuzyk, M. G., Dirk, C. W., Eds.;

Marcel-Dekker: New York, 1998.

(21) Polymers for Photonics Applications I: Nonlinear Optical and

Electro Luminescence Polymers; Advances in Polymer Science

158; Lee, K.-S., Ed.; Springer-Verlag: 2002.

(22) Robello, D. R.; Dao, P. T.; Schildkraut, J . S.; Scozzafava, M.;

Urankar, E. J .; Wiland, C. S. Chem. Mater. 1995, 7, 284-

291.

(23) Prime, R. B.; Chiou, G. Y.; Twieg, R. J . J . Therm. Anal. 1996,

46, 1133-1150.

Refer en ces a n d Notes

(1) (a) Dalton, L. Polym. Photonics Appl. I 2002, 158, 1-86. (b)

Eldada, L. Opt. Eng. 2001, 40, 1165-1178.

(2) (a) Hubbard, M. A.; Marks, T. J . Chem. Mater. 1992, 4, 965.

(b) Shi, Y.; Steier, W. H.; Chen, M.; Yu, L. P.; Dalton, L. R.

Appl. Phys. Lett. 1992, 60, 2577. (c) Wu, J . W.; Valley, J . F.;

Ermer, S.; Binkley, E. S.; Kenney, J . T.; Lipscomb, G. F.;

Lytel, R. Appl. Phys. Lett. 1991, 58, 225. (d) Mandal, B. K.;

Chen. Y. M.; Lee, J . Y.; Kumar, J .; Tripathy, S. Appl. Phys.

Lett. 1991, 58, 2549.

(24) Slessor, K. N. Can. J . Chem. 1971, 49, 2874.

(25) Robinson, B. H.; Dalton, L. R.; Harpe, A. W.; Ren, A.; Wang,

F.; Zhang, C.; Todorova, G.; Lee, M.; Aniszfeld, R.; Garner,

S.; Chen, A.; Steier, W. H.; Houbrecht, S.; Persoons, A.;

Ledoux, I.; Zyss, J .; J en, A. K. Y. Chem. Phys. 1999, 245, 35-

50.

(26) (a) Lopez, G.; Garcia, G.; Sanchez, G.; Garcia, J .; Ruiz, J .;

Hermoso, J . A.; Vegas, A.; Martinz-Ripoll, M. Inorg. Chem.

1992, 1, 1518. (b) De Haro, C.; Garcia, G.; Sanchez, G.; Lopez,

G. J . Chem. Res., Synop. 1986, 4, 119. (c) Falvello, L. R.;

Fornies, J .; Navarro, R.; Silicia, V.; Tomas, M. J . Chem. Soc.;

Dalton Trans. 1994, 12, 3143. (d) Falvello, L. R.; Fornies, J .;

Navarro, R.; Sicilia, V.; Tomas, M. Angew. Chem. 1990, 102,

952.

(3) Laschewsky, A.; Schulzhanke, W.; Makromol. Chem. Rapid

Commun. 1993, 14, 683-691.

(4) Grubbs, R. H.; Gorman, C. B.; Ginsburg, E. J .; Marder, S.

R.; Perry, J . W. Abstr. Pap. Am. Chem. Soc. 1990, 199, 345-

INOR, Part 1.

(5) (a) Maughon, B. R.; Weck, M.; Mohr, B.; Grubbs, R. H.

Macromolecules 1997, 30, 257. (b) Weck, M.; Mohr, B.

Maughon, B. R.; Grubbs, R. H. Macromolecules 1997, 30,

6430.

(6) (a) Lee, J .-K.; Schrock, R. R.; Baigent, D. R.; Friend, R. H.

Macromolecules 1995, 28, 1966. (b) Boyd, T. J .; Geerts, Y.;

Lee, J .-K.; Fogg, D. E.; Lovoie, G. G.; Schrock, R. R.

Macromolecules 1997, 30, 3553. (c) Bellmann, E.; Shaheen,

S. E.; Thayumanavan, S.; Barlow, S.; Grubbs, R. H.; Marder,

S.; Kippelen, B.; Peyghambarian, N. Chem. Mater. 1998, 10,

1668.

(27) Sekkat, Z.; Pretre, P.; Knoesen, A.; Volksen, W.; Lee, V. Y.;

Miller, R. D.; Wood, J .; Knoll. W. J . Opt. Soc. Am. B 1998

15, 401-413.

(7) Thorn-Csanyi, E.; Kraxner, P.; Strachota, A. Macromol. Rapid

Commun. 1998, 19, 223-228.

MA040044I

Article Doi

Synthesis and nonlinear-optical properties of vinyl-addition poly(norbornene)s

DOI: 10.1021/ma040044i

Source and publish data:

Authors:

Article abstract of DOI:10.1021/ma040044i

Full text of DOI:10.1021/ma040044i

Products guided by the article

R&D Labs maybe for 736981-09-2

Relevant to this article

Hot Product