Solvatochromic properties of long alkyl chain π* indicators: Comparison of N/N-dialkyl-4-nitroanilines and alkyl 4-nitrophenyl ethers

Article abstract of DOI:10.1002/poc.1148

Hydrophobic forms of the N/N-dialkyl-4-nitroaniline (DNAP) (p-O ₂NC₆H₄NR₂) (1a-f) and alkyl-4-nitrophenyl ether (p-O₂NC₆H₄OR) (2a-c) solvatochromic π* indicators have been charac

Full text of DOI:10.1002/poc.1148

JOURNAL OF PHYSICAL ORGANIC CHEMISTRY
J. Phys. Org. Chem. 2007; 20: 321–331
Published online 8 April 2007 in Wiley InterScience
(www.interscience.wiley.com) DOI: 10.1002/poc.1148
Solvatochromic properties of long alkyl chain pꢀ
indicators: comparison of N,N-dialkyl-4-nitroanilines
and alkyl 4-nitrophenyl ethers
1
1
1
2
2
2
2
*
R. Helburn, M. Bartoli, K. Pohaku, J. Maxka, D. Compton, B. Creedon and C. Stimpson
1Department of Chemistry and Physical Sciences, Pace University, 1 Pace Plaza, New York, New York 10038, USA
2Department of Chemistry, Northern Arizona University, Flagstaff, Arizona 86011, USA
Received: 18 November 2005; revised 5 July 2006; accepted 28 July 2006
ABSTRACT: Hydrophobic forms of the N,N-dialkyl-4-nitroaniline (DNAP) (p-O2NC6H4NR2) (1a–f) and alkyl-4-
nitrophenyl ether (p-O2NC6H4OR) (2a–c) solvatochromic pꢀ indicators have been characterized and compared with
ꢀ
respect to: (a) solvatochromic bandshape, (b) sensitivity expressed as Ss, (dvmax/dp ), and (c) trends in ꢁs with
increasing length of alkyl chain(s) on the probe molecule. —Octyl 4-nitrophenyl ether (p-O2NC6H4OC8H17) (2b) and
—decyl 4-nitrophenyl ether (p-O2N C6H4 OC10H21) (2c) were synthesized and their solvatochromic UV/Vis
absorption bands were found to maintain a Gausso-Lorentzian bandshape for the indicators in non-polar and alkyl
substituted aromatic solvents, for example, hexane(s) and mesitylene. Corresponding absorption bands for 1a–f
display increasing deviation from a Gausso-Lorentzian shape in the same solvents as the alkyl chains on the indicator
are increased in length all the way to C10 and C12, for example, N,N-didecyl-4-nitroaniline (p-O2NC6H4N (C10H21)2)
and N,N-didodecyl-4-nitroaniline (p-O2NC6H4N (C12H25)2) (1d–f). A plot of ꢁs versus Cn follows a 1st order decay
for the DNAP indicators but is linear for the alkyl 4-nitrophenyl ethers. A discussion of how the long alkyl chains on
the two types of indicators affect the orientation and overlap of n and pꢀ orbitals, and resulting solvatochromic bands is
presented. For DNAP, overextending the alkyl chains to obtain greater hydrophobic character may cause the alkane
component to dominate solute-solvation processes at the expense of the probe’s fundamental solvatochromic
character. Copyright # 2007 John Wiley & Sons, Ltd.
KEYWORDS: polarity; solvatochromism; indicator; solvenT; pꢀ
INTRODUCTION
phobic forms that we have prepared. They are the long
chain N,N-dialkyl-4-nitroanilines and some long chain
alkyl 4-nitrophenyl ethers (Figs 1 and 2). These newer
dyes comprise lipophilic forms of two of the seven
original indicators employed by Kamlet et al. in their
development of the original pꢀ scale.1
The pꢀ scale of solvent dipolarity and polarizability
which is based on the UV/Vis spectral shifts of individual
indicator solutes has found numerous uses since it was
first introduced in 1977 by Kamlet, Abboud, and Taft.1–2
Subsequent modification to scale parameters3 and to the
solvatochromic indicators used to produce individual
dipolarity measurements, pꢀ values,4–6 attest to the
interest that the pꢀ scale has generated and to its overall
broad utility. Among the more recent developments is the
synthesis and application of indicators with increasingly
long hydrophobic tails that permit the probing of solvent
environments inside micelles and lipid bilayers, thus
extending pꢀ measurements to these more complex
organic interfacial systems for which there is consider-
able interest.6–11 In this work, we present and compare the
solvatochromic properties of two of the more hydro-
Of particular interest in the characterization of new
polarity-sensitive indicators are (1) the shape of an
individual spectral band from which the wavenumber of
maximum absorption ðvmaxÞ is determined and (2) the
magnitude of the slope of a linear plot of vmaxversus
solvent pꢀ, the ꢁs parameter.1 Deviation from a Gaussian
or Lorentzian band shape can result in a lack of accuracy
and precision in locating vmax and in certain cases may
invalidate the use of the indicator as a polarity probe. ꢁs
provides a measure of an indicator’s sensitivity, for
example, spectral shifting capability. Larger values (of
ꢁs) result in a greater ability of the probe to resolve small
differences in dipolarity among solvation environments.
For heterogeneous systems where the probe resides in an
anisotropic environment, such differences can be signifi-
cant. Optimization of these two properties is thus a
*Correspondence to: R. Helburn, Department of Chemistry and
Physical Sciences, Pace University, 1 Pace Plaza, New York, NY
10038, USA.
E-mail: rhelburn@fsmail.pace.edu
Copyright # 2007 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2007; 20: 321–331

322
R. HELBURN ET AL.
R ..
R
In Equation 1, vmax is the wavenumber of maximum
absorption, in cmꢁ1, for the indicator in the solvent of
interest; vo is the wavenumber of maximum absorption
for the indicator in cyclohexane. Within the context of the
pꢀ scale, the slope s (Eqn 1) reflects the magnitude of
spectral shift over a range of solvents that span two
reference points, cyclohexane (pꢀ ¼ 0.00) and dimethyl-
sulfoxide (pꢀ ¼ 1.00).1 A correction term (dd) that weighs
the relative polarizability of individual solvents according
to membership in one of three structural classes, for
example, chlorinated (dd ¼ 0.5), aliphatic (dd ¼ 0.0), or
aromatic (dd ¼ 1.00) solvents, was added to the LSER
(Eqn 1) in 19813 though the parameter has not been
consistently applied.
N
+
N
O
O-
R= alkyl groups, each
with same no. of carbons
(n); n= 1 (1a) , 2 (1b), 3 (1c),
8 (1d), 10 (1e), 12 (1f).
Figure 1. Homologous series of di-n-(alkyl)-4-nitroaniline
(DNAP) indicators: N,N-dimethyl-4-nitroaniline (1a), N,N-
diethyl-4-nitroaniline (1b), N,N-dipropyl-4-nitroaniline (1c),
N,N-dioctyl-4-nitroaniline (1d), N,N-didecyl-4-nitroaniline
(1e), N,N-didodecyl-4-nitroaniline (1f)
The first pꢀ scale1 was based on the UV/Vis spectra of
seven indicators where the intent was to create an
approach that combined the individual spectral anomalies
of a suite of probes.1 It was later suggested that such
averaging of spectral information blurred meaningful
contributions and that a scale based on the UV/Vis
absorption bands of a single indicator was to be
preferred.2 Of the seven initial probes, the N,N-
dialkyl-4-nitroanilines, 1a and 1b, were found to be
most sensitive to non-specific solvent interactions (e.g.,
high ꢁs, Eqn 1). However, 1b was reported to have a
significant solvent-dependent bandshape suggesting that
longer-chain versions of this indicator might suffer
similarly. 4-Nitroanisole (2a), while less sensitive, that
is, lower ꢁs, was considered to be most desirable based
on its Gausso-Lorentzian bandshape which was seen to be
constant for media ranging all the way from the gas phase
to the most polar solvents.2,12 As we have prepared long
alkyl chain versions of both 1a–b and 2a (Figs 1 and
2),13,14 we can now pursue a more rigorous comparison of
the solvatochromic and spectral band shape properties of
these more hydrophobic pꢀ indicators.
consideration in tailoring new pꢀ indicators for use in
these complex systems.
Theory and background
Solvatochromic dyes act as indicators of solvent polarity
by exhibiting shifts in the positions of their UV/Vis
absorption bands. The ‘push–pull’ molecular system
illustrated in Fig. 3 shows the structural features that give
solvatochromic dyes their unique spectroscopic proper-
ties. The pꢀ indicators are a class of solvatochromic dyes
that respond to the non-specific portion of van der Waal’s
forces, for example, solvent induction, dispersion, and
dipolar effects,1–2 where the pꢀ parameter is calculated
from the positions of UV/Vis absorption bands of an
indicator in the solvent environment of interest. Exper-
imental values (of pꢀ) are related to the transition energy
through a linear solvation energy relationship (LSER) of
the following form.
EXPERIMENTAL
vmax ¼ vo þ spꢀ
(1)
Reagents (solvents)
Acetonitrile (99.9% anhydrous), benzonitrile (99þ%
spectrophotometric grade), cyclohexane (99þ% spectro-
photometric grade), hexanes and n-hexane (spectro-
photometric grade), n-heptane (spectrophotmetric grade),
n-pentane (99þ% spectrophotometric grade), 1,4-dioxane
(99.0% spectrophotometric grade), N,N-dimethylaceta-
mide (99þ% spectrophotometric grade), N,N- dimethyl-
formamide (DMF)(99.8% anhydrous), dimethylsulfoxide
(DMSO) (99.7% anhydrous), chlorobenzene (99þ%
spectrophotometric grade), benzene (99þ% spectro-
photometric grade), toluene (99.8% anhydrous), mesity-
lene (99%), —butyl acetate (spectrophotometric grade),
ethyl acetate (99.9% spectrophotometric grade),
diethyl ether(anhydrous), di-butyl ether (99þ%),
di-2-propyl ether (99% anhydrous), tetrahydrofuran
O
O
O
N
+
+
+
N
N
O-
O-
O-
O
O
O
2a.
2c.
2b.
Figure 2. Homologousseriesofsolvatochromic4-alkyloxynitro-
benzenes: 4-nitroanisole (2a), 4-octyloxynitrobenzene (2b),
and 4-decyloxynitrobenzene (2c)
Copyright # 2007 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2007; 20: 321–331
DOI: 10.1002/poc

SOLVATOCHROMIC PROPERTIES OF ANILINES
323
R
R
Electron-Pair-Donor
. .
R
R
O
R
R
+
O
N
+
O
N
Conjugated
System
+
+
+
+
N
N
N
N
O-
O
O-
O
O-
O-
-
-
O
Electron-Pair-Acceptor
e.
b.
c.
d.
a.
Figure 3. Model of a solvatochromic pꢀ indicator (a); dipolar mesomeric structures of alkyl-4-nitrophenyl ethers in the ground
(b) and excited (c) state; dipolar mesomeric structures of DNAP indicators in the ground (d) and excited (e) state
(THF) (99.5% anhydrous), pyridine (99.8% anhydrous),
N-methylpyrrolidin-2-one (NMP), 1,2-dichloroethane
(extra dry), tetrachloroethene (99þ% spectrophotometric
grade), and triethylamine (99.7%) were purchased from
Acros and used as received.
detector. Molecular masses were taken from the mass
spectral data.
Synthesis
An homologous suite of N,N-dialkyl-4-nitroanilines was
synthesized using methods described in Mansour et al.5
Indicators in this group that were characterized for the
present study are the crystalline N,N-dipropyl-4-nitroani-
line, N,N-dioctyl-4-nitroaniline, N,N-didecyl-4-nitroaniline,
and N,N-didodecyl-4-nitroaniline. 4-Alkyloxynitrobenzenes
with alkyl chain lengths of two, four, six, seven, eight, and
ten carbons were synthesized as follows.
Instrumentation
UV/Vis spectra were recorded on both Varian Cary 3E
and Beckman DU-640 scanning dispersive UV/Vis
spectrophotometers. Values for spectral bandwidth were
1.2 and 1.8 nm, respectively. Spectral data collection
intervals were 0.4 and 0.5 nm, respectively. For the Cary
3E, the cell block was thermostatted to 25 8C. Ambient
room temperatures were used in the operation of the
Beckman DU-640. All spectra were digitized and
converted to an ASCII format and imported into a
Quattro Pro or Excel spreadsheet for background
correction and subsequent data analysis.
1H NMR spectra, obtained for structure confirmation
on the synthesized alkyl 4-nitrophenyl ethers, were measured
at 400 MHz on Gemini Varian FT-NMR instrumentation.
Chemical shifts were referenced to tetramethylsilane
(TMS). CDCl3 was used as the supporting solvent in all
NMR measurements. Separations for synthesis reaction
mixtures were carried out using thin-layer chromatog-
raphy (TLC) and gas chromatography with mass
spectrometric detection (GC/MS). Diagnostic TLC was
performed using Whatman 250 mm silica gel plates. GC/
MS data were collected on an HP Model 5890
chromatograph with an HP Model 5971 mass selective
Procedure for synthesis of alkyl 4-nitrophenyl
ethers (R—O—Ph—NO2). In a 3-necked 250-mL
round-bottomed flask equipped with a dropping funnel
and reflux condenser, approximately 0.02 mole of
4-nitrophenol was added with approximately 0.02 mole
of K2CO3 in 40 mL of acetone. The acetone solution was
brought to reflux and neat haloalkanes were added slowly.
The reaction was followed by TLC until completion about
48–72 h, depending on alkyl chain length. After cooling,
the acetone was removed in a rotary evaporator. The solid
was dissolved in 30 mL of distilled water and extracted
with three 30 mL portions of tert-butyl methyl ether. The
ether solution was washed with three 30 mL portions of
3 M aqueous NaOH solution until the yellow color of the
phenolate anion was no longer present. The ether was
dried and then removed by rotary evaporation. The
materials are low melting solids. No melting points were
recorded. Purity was checked by GC/MS and NMR. The
yields were not optimized (Table 1).15
Table 1. Summary of procedure for synthesis of 2b–c and additional species
R-
K2CO3 mmol (g) 4-nitrophenol mmol (g)
R–X mmol (g)
Reflux time (h) R–O–Ph–NO2 (g) % yield
-Butylb
-Hexyl
-Heptylb
-Octyla
-Decyla
17.03 (2.34)
19.58 (2.70)
18.59 (2.57)
18.86 (2.60)
18.50 (2.55)
16.84 (2.343)
20.23 (2.814)
19.18 (2.669)
18.07 (2.516)
18.999 (2.643)
Iodobutane 16.90 (3.11)
Iodohexane 19.64 (4.16)
Iodoheptane 18.30 (4.13)
Iodooctane 16.69 (4.00)
Iododecane 18.35 (4.92)
48
57
64
70
71
(2.36) 71%
(1.86) 43%
(2.58) 63%
(2.90) 69%
(3.34) 61%
a Long alkyl chain dyes used in spectral studies and comprehensive ꢁs determinations.
b Used to corroborate observed trends in ꢁs.
Copyright # 2007 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2007; 20: 321–331
DOI: 10.1002/poc

324
R. HELBURN ET AL.
Identification of the alkyl 4-nitrophenyl ethers
(R—O—Ph—NO2). Butyl 4-nitrophenyl ether (CAS
7244-78-2). 1H NMR (CDCl3) d 0.98 (t 3H, CH3) 1.5 (m
2H, CH2), 1.8 (m 2H, CH2), 4.05 (t 2H, CH2), 6.93 (d 2H,
Ar), 8.16 (d 2H, Ar); GC/MS m/z 195, calcd m/z
for C10H13NO3 195.
2H, CH2), 1.78 (m 2H, CH2), 3.98 (t 2H, CH2), 6.87 (d
2H, Ar), 8.09 (d 2H, Ar); GC/MS m/z 279, calcd m/z
for C16H25NO3 279.
Determination of Ss and bandshape studies
Hexyl 4-nitrophenyl ether (CAS 15440-98-9). 1H NMR
(CDCl3) d 0.89 (t 3H, CH3), 1.33 (m 6H CH2), 1.45 (m
2H, CH2), 1.81 (m 2H, CH2), 4.03 (t 2H, CH2), 6.91 (d
2H, Ar), 8.15 (d 2H, Ar); GC/MS m/z 223, calcd m/z
for C12H15NO3 223.
Values of ꢁs were calculated for each new indicator from
their individual UV/Vis spectra in 20–23 non-
hydrogen-bond-donor (non-HBD) solvents. The solvents
(Tables 2–4) were selected from a list of non-HBD and
non-hydrogen-bond-acceptor (non-HBA) solvents uti-
lized by Kamlet et al.1 in the construction of their original
pꢀ scale and in the estimation of ꢁs for several of their
original small molecule indicators.1 Values of vmax were
determined from the 90% method16 and from the best fit
Gaussian or Lorentzian as applied to the top 0.1 to 0.15
absorbance range of an individual band; note: our
preliminary studies showed that fitting of the entire
spectral band yields lmax values that are affected by the
adjacent baseline.13 The ‘goodness’ of the fit, for
example, Gaussian or Lorentzian, is expressed via
Chi-square and r2 values, obtained from the individual
fitting routines. Model fitting was implemented using
Origin ver. 6.1 software. In addition to providing vmax
values, Gaussian and Lorentzian fitting of individual
spectra was used to compare bandshapes for C8 and C10
Heptyl 4-nitrophenyl ether (CAS 13565-36-1).
1H NMR (CDCl3) d 0.89 (t 3H, CH3), 1.32 (m
6H CH2), 1.45 (m 2H, CH2), 1.81 (m 2H, CH2), 4.03
(t 2H, CH2), 6.91 (d 2H, Ar), 8.15 (d 2H, Ar); GC/MS m/z
237, calcd m/z for C13H19NO3 237.
Octyl 4-nitrophenyl ether (CAS 49562-76-7). 1H NMR
(CDCl3) d 0.88 (t 3H, CH3), 1.31 (m 8H, CH2), 1.46 (m
2H, CH2), 1.81 (m 2H, CH2), 4.04 (t 2H, CH2), 6.93 (d
2H, Ar), 8.19 (d 2H, Ar); GC/MS m/z 251, calcd m/z
for C14H21NO3 251.
Decyl 4-nitrophenyl ether (CAS 31657-37-1). 1H NMR
(CDCl3) d 0.84 (t 3H, CH3), 1.27 (m 12H, CH2), 1.41 (m
Table 2. Values of vmaxin cmꢁ1 for N,N-dipropyl-4-NO2 (1) and N,N-dioctyl-4-NO2 (2)
(1)
(2)
Solvent
Gaussiana
Lorentzianb
90/90
Gaussianc
Lorentziand
90/90
1
2
3
4
5
6
7
8
Benzonitrile
Chlorobenzene
n-Heptane
N,N-Dimethylacetamide
DMF
DMSO
24.767
25.385
27.543
24.709
24.662
24.309
25.853
25.978
26.829
26.793
25.669
24.767
24.590
24.963
27.070
26.661
25.818
26.621
27.373
27.719
24.908
26.231
24.765
25.384
27.543
24.707
24.660
24.307
25.851
25.976
26.828
26.792
25.669
24.765
24.588
24.962
27.070
26.660
25.817
26.621
27.373
27.718
24.885
26.230
24.69
25.32
27.55
24.7
24.625
25.279
27.369
24.707
24.624
25.279
27.369
24.707
24.61
25.27
27.35
24.7
24.66
24.32
25.71
25.91
26.72
26.72
25.65
24.77
24.58
24.95
27.06
26.63
25.78
26.58
27.38
27.68
24.89
26.24
24.223
25.783
25.867
26.693
26.653
25.581
24.702
24.521
24.828
26.910
26.535
25.718
26.388
27.209e
24.222
25.781
25.867
26.692
26.645
25.581
24.700
24.520
24.828
26.911
26.534
25.716
26.386
27.210e
24.2
Benzene
25.73
25.81
26.65
26.62
25.58
24.69
24.5
24.82
27.03
26.49
25.74
26.38
27.18e
n-Butyl acetate
Butyl ether
Di-isopropyl ether
THF
Pyridine
NMP
1,2-Dichloroethane
Triethylamine
Tetrachloroethene
1,4-Dioxane
Ethyl ether
n-Hexane
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
n-Pentane
Acetonitrile
Mesitylene
Ethyl acetate
26.227
25.740
26.231
25.740
26.24
25.71
a Range of Chi-square ¼ 9.5 ꢂ 10ꢁ6 to 4.0 ꢂ 10ꢁ5; range of r2 ¼ 0.9539 to 0.9886.
b Range of Chi-square ¼ 9.3 ꢂ 10ꢁ6 to 4.0 ꢂ 10ꢁ5; range of r2 ¼ 0.9489 to 0.9893.
c Range of Chi-square ¼ 7.10 ꢂ 10ꢁ6 to 5.0 ꢂ 10ꢁ5; range of r2 ¼ 0.9037 to 0.9923.
d Range of Chi-square ¼ 7.15 ꢂ 10ꢁ6 to 6.0 ꢂ 10ꢁ5; range of r2 ¼ 0.9152 to 0.9919.
e Values are for cyclohexane.
Copyright # 2007 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2007; 20: 321–331
DOI: 10.1002/poc

SOLVATOCHROMIC PROPERTIES OF ANILINES
325
Table 3. Values of vmax in cmꢁ1for N,N-didecyl-4-NO2 (1) and N,N-didodecyl-4-NO2 (2)
(1)
(2)
Solvent
Gaussiana
Lorentzianb
90/90
Gaussianc
Lorentziand
90/90
1
2
3
4
5
6
7
8
Acetone
25.073
24.597
25.272
27.359
27.199
24.617
24.551
24.204
25.813
26.008
25.871
25.717
26.454
26.670
26.651
25.574
24.679
24.474
24.844
26.866
26.524
25.073
24.596
25.272
27.359
27.199
24.616
24.548
24.203
25.811
26.006
25.870
25.716
26.453
26.670
26.649
25.573
24.679
24.478
24.844
26.866
26.525
25.14
24.58
25.25
27.33
27.21
24.61
24.54
24.2
25.63
25.98
25.87
25.73
26.46
26.64
26.62
25.54
24.67
24.45
24.83
27.01
26.50
25.080
24.624
25.290
27.354
27.196e
24.642
24.554
24.199
25.825
25.992
25.859
25.730
25.078
24.615
25.290
27.354
27.196e
24.640
24.553
24.198
25.823
25.997
25.823
25.997
25.06
24.62
25.29
27.34
27.26e
24.66
24.55
24.2
25.82
26.028
25.85
25.72
Benzonitrile
Chlorobenzene
n-Heptane
n-Hexane
N,N-Dimethylacetamide
DMF
DMSO
Benzene
Toluene
n-Butyl acetate
Ethyl acetate
di-ethyl ether
di-butyl ether
di-2-propyl ether
THF
Pyridine
NMP
1,2-Dichloroethane
Triethylamine
Tetrachloroethene
Mesitylene
9
10
11
12
13
14
15
16
17
18
19
20
21
22
26.715
25.720f
25.590
24.629
24.496
24.871
26.930
26.570
26.254
26.714
25.720f
24.628
24.494
24.494
24.876
26.931
26.571
26.252
26.64
25.71f
25.55
24.67
24.5
24.85
26.92
26.5
26.25
a Range of Chi-square ¼ 9.5 ꢂ 10ꢁ6 to 4.0 ꢂ 10ꢁ5; range of r2 ¼ 0.9539 to 0.9886.
b Range of Chi-square ¼ 9.3 ꢂ 10ꢁ6 to 4.5 ꢂ 10ꢁ5; range of r2 ¼ 0.9489 to 0.9893.
c Range of Chi-square ¼ 7.10 ꢂ 10ꢁ6 to 5.0 ꢂ 10ꢁ5; range of r2 ¼ 0.9037 to 0.9923.
d Range of Chi-square ¼ 7.15 ꢂ 10ꢁ6 to 6.0 ꢂ 10ꢁ5; Range of r2 ¼ 0.9152 to 0.9919.
e Values are for cyclohexane.
f Values are for the cyclic ether dioxane.
4-alkyloxynitrobenzenes with those of the C3, C8, C10,
and C12 DNAP indicators.
(Eqn 1) for 2b and 2c (Table 4). We note that the spectrum
of pyridine overlaps too strongly with the solvatochromic
band of 2a in that solvent while 2b and 2c have
solvatochromic bands that are slightly red shifted when in
pyridine such that this solvent is usable for establishing pꢀ
scales based on the longer-chain alkyloxy dye.
RESULTS AND DISCUSSION
Solvents and solvatochromic
absorption bands
Values of vmax obtained via the three methods,
Gausssian and Lorenztian fit, and the 90% method of
Kamlet et al.,16 from individual UV/Vis spectra of the
N,N-dipropyl, dioctyl, didecyl, and didodecyl-4-
nitroanilines in each of 20–23 solvents are given in
Tables 2–3. Corresponding vmaxvalues for the C8 and C10
alkyl 4-nitrophenyl ethers are listed in Table 4. Use of
both Gaussian and Lorentzian spectral band fitting as a
means for locating and estimating lmax for bands of these
two types of pꢀ indicators has permitted us to
simultaneously examine band shape in a statistical
manner, for example, as the extent of deviation from a
Gaussian or Lorentzian bandshape or by comparing lmax
from the individual routines. For example, solvatochro-
mic bands of the DNAP indicators in all solvents showed
differences in magnitude of fitted lmax, between
individual Gaussian and Lorentzian fit, that ranged from
0 to 0.37 nm. The differences between the two approaches
were not more than 0.03 nm for bands of the C8 and C10
alkyl 4-nitrophenyl ethers. The errors in individual
The ranges of lmax for solvatochromic bands of the
DNAP indicators (Fig. 1) versus those of the alkyl
4-nitrophenyl ethers (Fig. 2) in non-HBD solvents of
varying dipolarity (Tables 2–4) are 350–424 nm and
292–319 nm, respectively. The more blue-shifted range of
lmax for the alkyl nitrophenyl ethers resulted in our
excluding acetone and other ketones from the solvent list
for these indicators since UVabsorption by those solvents
overlapped strongly with the solvatochromic absorption
bands of 2b and 2c, which accordingly could not be
resolved. Spectra of other conjugated solvents such as
benzene, toluene, mesitylene, and NMP overlap partially
with solvatochromic bands of the long chain alkyloxy
dyes such that bands for these indicators could be
resolved when the solvent spectrum was subtracted from
that of the solution; thus these latter solvents were
included in the list of those used in establishing LSERs
Copyright # 2007 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2007; 20: 321–331
DOI: 10.1002/poc

326
R. HELBURN ET AL.
Table 4. Values of vmax in cmꢁ1 for 4-octyloxynitrobenzene (1) and 4-decyloxynitrobenzene (2)
(1)
(2)
#
Solvent
Gaussiana
Lorentzianb
Gaussianc
Lorentziand
1
2
3
4
5
6
7
8
Acetonitrile
Benzonitrile
Chlorobenzene
Cyclohexane
n-Heptane
N,N-Dimethylacetamide
DMF
DMSO
32.287
31.760
32.220
33.790
33.931
31.809
31.800
31.323
32.521
32.886
32.751
33.298
33.364
32.510
31.735
31.560
31.815
33.667
33.184
32.599
32.287
31.758
32.219
33.789
33.930
31.809
31.930
31.323
32.521
32.883
32.751
33.296
33.363
32.510
31.736
31.560
31.814
33.666
33.184
32.598
32.184
31.706
32.201
34.071
33.940
31.820
31.774
31.480
32.503
32.753
32.818
33.283
33.367
33.155
31.709
31.576
31.984
33.765
33.135
32.639
34.211
33.374
32.592
32.184
31.706
32.501
34.069
33.939
31.819
31.773
31.478
31.501
32.752
32.817
33.283
33.367
33.153
31.707
31.576
31.983
33.764
33.136
32.639
34.214
33.374
32.592
9
Benzene
10
11
12
13
14
15
16
17
18
19
20
21
22
23
Mesitylene
Ethyl acetate
di-ethyl ether
di-butyl ether
THF
Pyridine
NMP
1,2-Dichloroethane
Triethylamine
Tetrachloroethene
Dioxane
Pentane
di-2-propyl ether
Toluene
33.35
32.61
33.46
32.59
a Range of Chi-square ¼ 1.85 ꢂ 10ꢁ6 to 2.0 ꢂ 10ꢁ5; single outliers at 1.8 ꢂ 10ꢁ4 and 3.8 ꢂ 10ꢁ4 for benzonitrile and triethylamine, respectively; range of
r2 ¼ 0.9745 to 0.9982.
b Range of Chi-square ¼ 2.02 ꢂ 10ꢁ6 to 2.0 ꢂ 10ꢁ5; single outliers at 1.9 ꢂ 10ꢁ4 and 3.9 ꢂ 10ꢁ4 for benzonitrile and triethylamine, respectively; range of
r2 ¼ 0.9739 to 0.9980.
c Range of Chi-square ¼ 7.78 ꢂ 10ꢁ7 to 1.0 ꢂ 10ꢁ5; range of r2 ¼ 0.9767 to 0.9986.
d Range of Chi-square ¼ 1.37 ꢂ 10ꢁ6 to 2.0 ꢂ 10ꢁ5; range of r2 ¼ 0.9870 to 0.9985.
spectral fitting routines expressed as Chi-square (note: the
larger the Chi-square value, the poorer the fit) and r2 are
also relevant. For example, the average Chi-square value
for individual Gaussian fits over the range of solvents was
36 times greater for 1e (N,N-didecyl-4-nitroaniline) than
for the single chain 4-decyloxynitrobenzene (2c). Look-
ing at the corresponding r2, we see that 90% of the r2
values were 0.99 or greater for 2c, whereas for 1e only 5%
of those values were greater than 0.99. A very similar
result was obtained for the Lorentzian fits where average
Chi–square values for fitting solvatochromic bands of
N,N-didecyl-4-nitroaniline (1e) were 28 times greater
than those of 2c (4-decyloxynitrobenzene) with 90% of r2
greater than 0.99 for the latter alkyloxy dye and just 5% of
r2 greater than 0.99 for the N,N-didecyl-4-nitroaniline.
The implications of these fitting data (e.g., summaries
of r2 and Chi-square for Gaussian and Lorentzian fits) are
illustrated in Figs 4 and 5. Figure 4 shows that both
solvatochromic bands of 2c and its short alkyl chain
counterpart 2a in the non-polar solvent hexanes maintain
a Gausso-Lorentzian bandshape. Likewise, the solvato-
chromic band for 1e in hexanes deviates in shape from
that of 1a, presenting an overall less ‘Gaussian-like’ band.
Laurence et al. (1994)2 have stated that of all the smaller
molecule solvatochromic pꢀ indicators, 1a and 2a are
among the choice molecular probes based on their
constant Gausso-Lorentzian bandshape which is main-
tained over a wide range of solvent polarities, but that
lengthening the alkyl chain on 1a to two carbons, for
example, 1b, results in severe bandshape distortions in
spectra of 1b, especially for the most non-polar alkane
solvent environments. Unlike that of the DNAP indicators
(Fig. 5), we observe that long alkyl chain versions of
4-nitroanisole, 2a, do not exhibit such spectral bandshape
deviations (Fig. 4).
In the earlier studies of indicator-specific spectral
anomalies,12 it was further suggested that polarizable
Figure 4. Solvatochromic (UV/Vis) absorption bands of 2a
(left) and 2c (right) in hexane(s)
Copyright # 2007 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2007; 20: 321–331
DOI: 10.1002/poc

SOLVATOCHROMIC PROPERTIES OF ANILINES
327
is highlighted for 1b (Fig. 6–1b) where these results are
consistent with previous findings.12 The distortion
becomes increasingly exaggerated for 1c and 1d such
that a dual peak spectral fitting routine, as opposed to a
single Gaussian or Lorentzian, more accurately describes
the solvatochromic band. Values of r2 and Chi-square for
the dual peak fit of the band(s) of 1d are 0.99 and
4.6 ꢂ 10ꢁ6, respectively. Conversely, the spectrum of 2b,
the corresponding alkyloxy indicator, in mesitylene
(Fig. 7) shows that the long chain alkyl 4-nitrophenyl
ethers are not similarly affected; the solvatochromic band
for 2b, 4-octyloxynitrobenzene, in mesitylene shows no
such spectral distortion; r2 ¼ 0.98, Chi-square ¼ 5.8 ꢂ
10ꢁ6 for the single peak Gaussian fitted band.
Figure 5. Solvatochromic (UV/Vis) absorption bands for 1a
and 1e in hexane(s)
solvents such as toluene, p-xylene, and mesitylene could
also influence the vibrational fine structure and bandshape
for solvatochromic bands of 1b.12 In this work, we find
that such effects, while somewhat evident in spectra of 1b,
for mesitylene, become much further exaggerated in
spectra of 1d–e in that solvent. Fig. 6 illustrates the
progressive increase in deviation from a Gausso-
Lorentzian bandshape as the alkyl chains are increased
in length (1a–d) for selected DNAP indicators in
mesitylene. The first appearance of spectral broadening
Spectral effects and electron-pair-donor
groups
Earlier discussions of the effect(s) of different electron-
pair-donor groups on the UV/Vis spectroscopic properties
of pꢀ indicators have almost always invoked steric
attributes on the part of the donor substituent to explain
their differing effect on the solvatochromic properties of
the probe.1 In particular, it has been stated in the context
Figure 6. UV/Vis spectra of the upper portions of individual solvatochromic absorption bands (. . ...) with best fit single or
double peak Gaussian fit (___) for 1a–d in mesitylene. For Gaussian dual peak fit of 1d (N,N-dioctyl-4-nitroaniline): lmax
(1) ¼ 373.7 nm (outlier on LSER for 1d), lmax (2) ¼ 386.1 nm (point lies on the LSER for 1d; note: for single peak Gaussian fit of
spectrum d, lmax ¼ 381.2 nm
Copyright # 2007 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2007; 20: 321–331
DOI: 10.1002/poc

328
R. HELBURN ET AL.
compared to that of the alkyloxy dye, the result due to
motions of the alkyl chain pair is an uneven ability of the
(n) orbital to overlap directly with the neighboring pꢀ
orbital (Fig. 3e) resulting in two conformationally distinct
excited states, one that is less well relaxed (band at
lmax ¼ 373.7 nm, Fig. 6-1d) and one that is more stable
(band at lmax ¼ 386.1 nm, Fig. 6-1d), that is, based on
orientation of the n orbital with respect to the pꢀ orbital.
We suggest that the longer-wavelength lower-energy
absorption (lmax ¼ 386.1 nm, Fig. 6-1d) is the useful
solvatochromic band as vmax for this latter peak falls
neatly on the line of a linear plot of vmaxversuspꢀ while
vmax for the less stable excited state (lmax ¼ 373.7 nm,
Fig. 6d), which varies slightly more with alkyl chain
length than the longer wavelength band, exists as an
outlier on the LSER for that indicator. For
N,N-diethyl-4-nitroaniline (1b), this conformational
effect is not well pronounced resulting in an apparent
lmax for the solvatochromic band that is closer to an
average of these effects (Fig. 6-1b, lmax ¼ 381.2 nm).
The n orbital on the corresponding long chain
4-alkyloxynitrobenzene is oriented more directly toward
the solvent over a wide range of O-alkyl chain lengths.
Motions on the part of the single alkyl chain on these
nitrophenyl ether probes do not affect solvation of the n
electrons on the oxygen of this indicator or their
orientation with respect to the neighboring pꢀ orbital
on the excited-state resonance structure (see Fig. 3c).
Thus, no such conformational spectral effects are seen for
the probe in mesitylene (Fig. 7) or any other solvent.
Figure 7. UV/Vis spectrum of upper portion of solvatochro-
mic absorption band (. . .. . .) with best fit Gaussian (_____) of
2b in mesitylene
of n-alkyl groups that depending on structure, the donor
substituent can sterically hinder solvation of the
non-bonding (n) electron pair. This in turn affects
intramolecular charge transfer and the resulting dipolar
properties of the indicator in response to the solvent
(Fig. 3c and e). For small short-alkyl chain donor
substituents, the steric hindrance argument would seem
reasonable. But if a particular spectroscopic effect, for
example, bandshape distortion, becomes increasingly
pronounced as the alkyl chains are extended beyond that
which could be expected to provide simple steric blocking
of the lone electron-pair, then another mechanism might
be proposed. The two vibrationally separated absorption
bands (Fig. 6-1d) seen in spectra of N,N-dioctyl-4-
nitroaniline in mesitylene which are less pronounced for
the N,N-dipropyl (1c; Fig. 6-1c) and barely existent for
N,N-diethyl-4-nitroaniline(1b; Fig. 6-1b) are examples of
this situation. We suggest that motions of the longer alkyl
chains on the donor group affect the conformation of the
indicator and the orientation of the lone electron pair in
relation to the solvent and to the pꢀ orbital with which it
overlaps during a transition (n-pꢀ). Moreover, it is not
inconceivable that the very long alkyl chains affect the
nature of the solvent in the vicinity of the n orbital, for
these indicators. This latter effect is easy to envision in the
case of N,N-didecyl-4-nitroaniline (1e) and N,N-
didodecyl-4-nitroaniline (1f) as the alkyl moieties of
these dyes are so long that they actually become part of
the solvent in the n orbital region of the molecule.
That these spectral effects are seen primarily in pure
alkane solvents and in alkane substituted benzenes, both
of which have structural motifs in common with DNAP
indicators, suggest the existence of non-specific solute–
solvent interactions that are most pronounced for these
solvent–dye systems. Where the n orbital of the DNAP
indicator is not directly oriented toward the solvent as
LSERS and trends in Ss
LSERs (Eqn 1) were prepared for 1c–f and 2b–c using the
values of vmax listed in Tables 2–4, in order to rigorously
determine ꢁs for the probes and to observe trends in ꢁs
over the range of alkyl chain lengths for these two types of
pꢀ indicators (Figs 1 and 2). For consistency in
comparison of values, we have re-determined ꢁs for
the well characterized small molecule probes (1a–b and
2a) using the same solvent list (Tables 2–4); these values
compare favorably with those of Kamlet et al.1 Table 5
summarizes all ꢁs values for indicators in this work
computed to date, including some previously published
seven-point determinations.4,8
We suggest that the conformational effects proposed
for 1e–f in the solvent mesitylene, which resulted in an
outlier point for selected spectral bands (Fig. 6-1d) on
LSERs for those dyes, may also manifest themselves in
the region of the LSER that corresponds to vmax for linear
alkane solvents (pꢀ ꢃ 0.00). Our previous argument that
the -decyl and -dodecyl chains become part of the solvent
in the region of the n orbital for DNAP indicators may
explain the observed decrease in sensitivity of Dvmax as a
function of solvent pꢀ in the ‘alkane’ region of the LSER.
This phenomenon is best illustrated in the LSER for
Copyright # 2007 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2007; 20: 321–331
DOI: 10.1002/poc

SOLVATOCHROMIC PROPERTIES OF ANILINES
329
Table 5. Values of ꢁs, current and previous, for DNAP and 4-alkoxynitrobenzenes
Indicator
Method of lmax fit
# solvents
vo
ꢁs
References
1a (dimethyl)
90%
90%
90%
90%
Gaussian fit
Lorentzian fit
90%
90%
90%
7
7
28.18
28.23
28.10
28.18
28.17
28.17
27.52
27.52
27.52
27.53
27.55
27.55
27.41
27.39
27.39
27.43
27.43
27.18
27.30
27.30
27.30
27.21
27.26
27.25
27.25
27.32
27.33
27.35
34.17
34.16
34.16
33.91
33.91
33.95
33.95
3.445
3.464
3.436
3.44
Carrozzino et al. (2004)8
Helburn et al. (1997)4
Kamlet et al. (1977)1
This work
28
20
20
20
7
3.384
3.386
3.130
3.136
3.182
3.125
3.145
3.146
3.114
3.007
3.067
3.067
3.07
2.921
2.971
2.906
2.958
2.956
2.9608
2.927
2.928
2.997
3.017
3.04
This work
This work
1b (diethyl)
Carrozzino et al. (2004)8
Helburn et al. (1997)4
Kamlet et al. (1977)1
This work
7
28
20
20
20
7
90%
Gaussian fit
Lorentzian fit
90%
90%
90%
Gaussian fit
Lorentzian fit
90%
This work
This work
1c (di-propyl)
Carrozzino et al. (2004)8
Helburn et al. (1997)4
This work
7
21
21
21
7
20
20
20
7
20
20
20
20
20
20
28
21
21
22
22
23
23
This work
This work
1d (di-octyl)
1e (di-decyl)
Carrozzino et al. (2004)8
This work
This work
90%
Gaussian fit
Lorentzian fit
90%
This work
Carrozzino et al. (2004)8
This work
This work
90%
Gaussian fit
Lorentzian fit
90%
Gaussian
Lorentzian
90%
Gaussian
Lorentzian
Gaussian
Lorentzian
Gaussian
Lorentzian
This work
This work
This work
This work
1f (di-dodecyl)
2a (methoxy)
2.410
2.409
2.405
2.433
2.425
2.436
2.415
Kamlet et al. (1977)1
This work
This work
2b (-octyloxy)
2c (-decyloxy)
This work
This work
This work
This work
N,N-didodecyl-4-nitroaniline (Fig. 8). As the polarity of
the solvent approaches that of a linear alkane (pꢀ < 0.00),
solute–solvent interactions which are fewer than those for
the more dipolar solvents become dominated by the long
alkyl chains on the indicator, which is now more
alkyl-like than phenyl-like, such that the solvent in the
region of the n electrons becomes essentially unchanging
and unable to elicit any further shift in the dipolar nature
of the probe molecule. The result is an overall decrease in
sensitivity of vmax to changes in solvent pꢀ in the negative
pꢀ region of the plot (Fig. 8). It is of interest to note that
the organization of alkyl chains for DNAP indicators in
the context of their crystal structure also is different for
the short and long chain species. In a previous paper,5 we
reported that the layered packing of 1c (N,N-
dipropyl-4-nitroaniline) lacks a center of symmetry
resulting in a polar crystal, while layers of N,N-
didecyl-4-nitroaniline (1d) exhibit cosymmetry such that
the crystal is non-polar.5
Figure 8. LSER (vmaxvs. pꢀ, Eqn. 1) for N,N-didodecyl-
4-nitroaniline. Values of vmax determined by Lorentzian
fitting of the upper portion of individual indicator bands.
Linear fit (__) r2 ¼ 0.9730; polynomial fit (----) r2 ¼ 0.9843;
The observed change in ꢁs over the range of
carbon-chain length illustrated for both DNAP and
nitrophenyl ether indicators (Fig. 9) is in support of the
Copyright # 2007 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2007; 20: 321–331
DOI: 10.1002/poc

330
R. HELBURN ET AL.
that of the lone pair on the alkyloxy dye (Fig. 2). The latter
is more evenly exposed to the solvent over a wide range of
alkyl chain lengths on the OR substituent. As a result, ꢁs
for the latter alkyloxy suite is essentially constant and
linear over the range of methoxy, octyloxy, and decyloxy
species (Fig. 9b). Note that we have inserted some
two-point determinations of ꢁs, based on vmax for
indicators in DMSO and cyclohexane only, for the C2, C4,
and C7 alkyloxy indicators to confirm the linearity of the
plot (Fig. 9b).
Hydrophobic pꢀ indicators, past
and current effort
In an earlier paper of this type,4 we made a strong
argument for the N,N-dialkyl-4-nitroanilines as candidate
structures for more hydrophobic versions of the pꢀ
indicators1 based on their relatively large values of ꢁs and
observed constancy of ꢁs among some of the longer alkyl
chain probes.4 However, an equally strong argument can
be made for the 4-alkyloxy nitrobenzenes on the basis of
their ‘less perturbing’ single alkyl chain structure, more
constant bandshape, and the constancy of ꢁs over the
entire range of carbon-chain lengths. While the pheny-
lether dyes are slightly less sensitive to solvent dipolarity,
that is, ꢁs is smaller overall, this drawback may be more
than compensated by a spectroscopic behavior that is
stable over a wide range of solvent dipolarities and probe
alkyl chain lengths. We note that intercalation of probes
into lipid bilayers (a proposed application) may favor a
more slender single alkyl chain shape over the more bulky
dialkyl configuration. Overall, caution must be taken for
all pꢀ indicators in over extending the length of alkyl
chains such that gains made in generating hydrophobic
character for the molecule are not compromised by a loss
in their fundamental solvatochromic properties.
Figure 9. ꢁs versus # of carbons on alkyl chain. (a) Plot for
DNAP indicators including points obtained from vmax esti-
mated via the 90% (^) and Lorentzian fit (5); plot is fitted
with a 1st order exponential decay and (b) plot for
4-nitroalkoxybenzenes including points obtained from vmax
estimated via Gaussian band fit (&) and Lorentzian band fit
(*). Three center points (&) indicated by arrows represent
two-point estimates of ꢁs (based on vmax for cyclohexane
and DMSO only) for the C2, C4, and C7
4-alkyloxynitrobenzenes,18 these points were added to
confirm linearity of the plot
proposed mechanisms that we have discussed, though it
should be emphasized that s ðdvmax=dpꢀÞ represents an
average indicator response over many solvents rather than
specific spectral details (such as we have discussed in the
case of selected solvents). Overall, ꢁs is larger for the
DNAP indicators than for the nitrophenyl ether dyes
(Table 5) due, at least in part, to the stronger
electron-donating properties of the -NR2 dialkylamino
substituent over that of the -OR alkyloxy group, which
when paired with an electron-accepting -NO2 through the
conjugated system (Fig. 3), results in a stronger
intramolecular dipole (for DNAP) that is better aligned
with the solvent dipole, especially for the excited state
(Fig. 3e).17 Thus dDET/dpꢀ, which is directly proportional
to ꢁs, is larger and more pronounced for DNAP indicators
than for the alkyloxy dyes, with a slightly larger range of
spectral shift positions for the former. However, in the
case of the DNAP indicators, these effects vary with alkyl
chain length (Fig. 9a). The decrease in ꢁs on moving
from dimethyl (1a) to diethyl (1b) and the dipropyl (1c)
indicators can be explained by the corresponding increase
in shielding of the lone n electrons by the increasing bulk
and motions of the alkyl groups, an effect that levels off
with further increase in carbon number, forming a trend
that is best illustrated by a 1st order decay (Fig. 9a). The
orientation of the lone pair with respect to the two alkyl
chains, the neighboring pꢀ orbital, and the solvent for the
DNAP indicator (Fig. 1) is significantly different from
REFERENCES
1. Kamlet MJ, Abboud J-LM, Taft RW. J. Am. Chem. Soc. 1977; 99:
6027–6038.
2. Laurence C, Nicolet P, Tawfik-Dalati M, Abboud J-LM, Notario R.
J. Phys. Chem. 1994; 98: 5807–5816.
3. Taft RW, Abboud J-LM, Kamlet M. J. Am. Chem. Soc. 1981; 103:
1080–1086.
4. Helburn R, Ullah N, Mansour G, Maxka J. J. Phys. Org. Chem.
1997; 10: 42–48.
5. Mansour G, Creedon W, Dorrestein PC, Maxka J, Macdonald JC,
Helburn R. J. Org. Chem. 2001; 66: 4050–4054.
6. Helburn R, Dijiba Y, Mansour G, Maxka J. Langmuir 1998; 14:
7147–7154.
7. Vitha MF, Weckwerth JD, Odland K, Dema V, Carr P. J. Phys.
Chem. 1996; 100: 18823–18828.
8. Carrozzino JM, Fuguet E, Helburn R, Khaledi M. J. Biochem.
Biophys. Methods 2004; 60(2): 97–115.
9. Lu H, Rutan SC. Anal. Chem. 1996; 68(8): 1387–1393.
10. Helburn R, Rutan SC, Pompano J, Mitchem D, Patterson WT.
Anal. Chem. 1994; 66: 610–618.
Copyright # 2007 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2007; 20: 321–331
DOI: 10.1002/poc

SOLVATOCHROMIC PROPERTIES OF ANILINES
331
11. Shin DM, Schanze KS, Otruba JP, Brown PE, Whitten DG. Isr. J.
Chem. 1987/88; 28: 37–45.
12. Nicolet P, Laurence C. J. Chem Soc. Perkin Trans. 11 1986; 1071–
1079.
13. Creedon W. MS Thesis, Northern Arizona University, USA,
1999.
16. lmax is taken as the midpoint between wavelengths corresponding to
90% of the maximum absorbance of the solvatochromic band [1].
17. Nitroaromatic pꢀ indicators exhibit ‘positive’ solvatochromism,
i.e. spectral bands shift to longer wavelengths with increasing
solvent dipolarity, implying that the excited state of the dye is more
dipolar (in response to solvation) than the ground state.
14. Compton D, Maxka J, Helburn R. 4th Annual Meeting of the
Arizona-Nevada Academy of Science. Glendale, AZ, 1998.
15. Allen CFH, Gates JW , Jr. Org. Synth. Coll. 1955; 3: 140–141.
18. C2, C4, and C7 alkyl-4-nitrophenyl ethers were synthesized using
methods as described. Only the longest alkyl chain species (C8
and C10) were fully characterized in 23 non-HBD solvents.
Copyright # 2007 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2007; 20: 321–331
DOI: 10.1002/poc