Detection of explosives via photolytic cleavage of nitroesters and nitramines

FEATURED ARTICLE

pubs.acs.org/joc

Detection of Explosives via Photolytic Cleavage of Nitroesters

and Nitramines

Trisha L. Andrew and Timothy M. Swager*

Department of Chemistry, Massachusetts Institute of Technology, 77 Massachusetts Avenue, Cambridge, Massachusetts 02139,

United States

S Supporting Information

b

ABSTRACT: The nitramine-containing explosive RDX and the

nitroester-containing explosive PETN are shown to be suscep-

tible to photofragmentation upon exposure to sunlight. Model

compounds containing nitroester and nitramine moieties are

also shown to fragment upon exposure to UV irradiation. The

products of this photofragmentation are reactive, electrophilic

NO_xspecies, such as nitrous and nitric acid, nitric oxide, and

nitrogen dioxide. N,N-Dimethylaniline is capable of being

nitrated by the reactive, electrophilic NO_xphotofragmentation

products of RDX and PETN. A series of 9,9-disubstituted 9,10-

dihydroacridines (DHAs) are synthesized from either N-phe-

nylanthranilic acid methyl ester or a diphenylamine derivative and are similarly shown to be rapidly nitrated by the

photofragmentation products of RDX and PETN. A new (turn-on) emission signal at 550 nm is observed upon nitration of

DHAs due tothegeneration ofﬂuorescentdonorꢀacceptor chromophores. Usingﬂuorescencespectroscopy, the presenceofca. 1.2 ng

of RDX and 320 pg of PETN can be detected by DHA indicators in the solid state upon exposure to sunlight. The nitration of

aromatic amines by the photofragmentation products of RDX and PETN is presented as a unique, highly selective detection

mechanism for nitroester- and nitramine-containing explosives and DHAs are presented as inexpensive and impermanent

ﬂuorogenic indicators for the selective, standoﬀ/remote identiﬁcation of RDX and PETN.

’ INTRODUCTION

Detecting hidden explosive devices in war zones and trans-

portation hubs is an important pursuit. The three most com-

monly used highly energetic compounds in explosive formu-

lations are 2,4,6-trinitrotoluene (TNT), 1,3,5-trinitrotriazinane

(RDX), and pentaerythritol tetranitrate (PETN) (Figure 1).

Numerous technologies are currently capable of detecting the

Figure 1. Structures of common high explosives.

energetic chemical components of explosive devices, including

analytical spot tests,¹ﬂuorescent sensors using either small-

molecule ﬂuorophores²or ﬂuorescent conjugated polymers,³

complex signal processing (i.e., ﬂuorescence lifetimes, depo-

chemiresistive sensors,⁴portable mass spectrometers,⁵and X-ray

larization).

systems.⁶Each example listed has unique advantages and limita-

We previously reported a turn-on ﬂuorescence chemosensing

tions. For instance, while X-ray systems are capable of detecting

bulk hidden explosive devices and portable mass spectrometers

are capable of identifying the exact chemical structures of suspect

chemicals, the practical deployment and/or longevity of these

hardware-intensive technologies in complex environments is

nontrivial.⁵Fluorescent sensors are comparatively technol-

ogy-unintensive, have desirably low detection limits, and are

also capable of identifying (responding to) entire classes of

molecules (such as nitroaromatics) or particular functional

groups (vide infra).³Chemical spot tests can be more speciﬁc

than ﬂuorescent sensors but are not as sensitive and do not

have the analytical advantages of an emissive signal, such as

remote line-of-sight (stand-oﬀ) detection or prospects for

scheme based on the photoreaction between a hydride donor and

either RDX or PETN, wherein the nitramine or nitroester

component was photoreduced by 9,10-dihydroacridine (AcrH₂,

Figure 2) or its metalated analogues.⁷The acridinium products

(AcrH^þ) of this photoreaction had a high ﬂuorescence quantum

yield and resulted in a signiﬁcant ﬂuorescence turn-on signal in

the presence of RDX and PETN.

While studying this photoreaction, we became interested in

the photochemical stability of nitramine and nitroester compounds

under ultraviolet (UV) irradiation. Nitroesters and nitramines

Received: February 5, 2011

Published: March 31, 2011

r

2011 American Chemical Society

2976

dx.doi.org/10.1021/jo200280c J. Org. Chem. 2011, 76, 2976–2993

|

The Journal of Organic Chemistry

FEATURED ARTICLE

colored azo dye (7).¹¹As seen in Scheme 1C, when nitrite test

strips impregnated with the modiﬁed Greiss reagent were dipped

into solutions of either RDX or PETN in 2:1 acetonitrile/1 M

NaOH, a bright pink color evolved, indicating the presence of

nitrite anions.

Interestingly, when the same nitrite test strips were dipped

into base-free acetonitrile solutions of RDX or PETN, dried, and

irradiated (λ = 254 nm), formation of the pink azo dye was also

observed (Scheme 1D), suggesting the evolution of nitrite ions

upon the photolysis of RDX and PETN. Photolysis at 313, 334,

and 365 nm similarly resulted in a positive Greiss test; however,

non-UV controls did not yield a pink color. Moreover, extended

exposure (30 min) to polychromatic light from a solar simulator

was also observed to photolyze RDX and PETN and yield a

positive Greiss test (Scheme 1E).

The photolysis of nitroester- and nitramine-based energetic

compounds under various conditions has been studied and found

to produce a number of small-molecule degradation products,

including nitrous and nitric acid, nitric oxide, nitrogen dioxide,

formaldehyde, and ammonia.¹²The proposed photolytic degra-

dation mechanisms for PETN and RDX are shown in Scheme 1F.

In the case of PETN, it is hypothesized that heterolytic cleavage

of the O-NO₂bond initially produces an alkoxide (8) and a

highly reactive nitronium ion (9) that rapidly forms nitric acid

under ambient conditions.¹³For RDX, evidence of both the

homolytic and heterolytic scission of the NꢀNO₂bond of RDX

Figure 2. Structures of the hydride donor AcrH₂, its oxidation product

AcrH^þ, and the 9,9-disubstituted 9,10-dihydroacridines, DHAs, studied

herein.

have been known to degrade under highly acidic or basic

conditions and established spot tests for PETN and RDX detect

these chemical degradation products as opposed to directly

detecting intact PETN or RDX.¹The base-promoted digestion

of nitroglycerin (NG) has also been studied and is thought to

evolve a mixture of nitrate and nitrite anions, among other

degradation products (Scheme 1A).⁸Similarly, RDX is also

known to decompose in basic media and produce nitrite ions

(Scheme 1A).⁹The Greiss test¹⁰for nitrite ions can, therefore, be

employed to conﬁrm the evolution of nitrite upon base-pro-

moted degradation of RDX and PETN. The chemistry behind

the commercially available Greiss test (Scheme 1B) involves the

reaction of sulfanilamide (4) with nitrite to form diazonium salt 6,

which then reacts with an arylamine (5) to form a brightly

Scheme 1. (A) Degradation Mechanisms of Nitroesters and Nitramines in Basic Media. (B) Active Components and Detection

Mechanism of the ZellerꢀGreiss Test for Nitrite Ions. (C) Nitrite Ion Test on Base-Degraded RDX and PETN.^a(D) Nitrite Ion

Test on Photolyzed RDX and PETN.^b(E) Nitrite Ion Test on RDX and PETN Exposed to Sunlight.^c(F) Proposed Photolytic

Cleavage Pathway of Nitroesters and Nitramines and Select Photofragmentation Products

a

Test strips were dipped into blank 2:1 MeCN/1 M NaOH (i) or 17 mg of PETN (ii) or 10 mg of RDX (iii) in 3 mL of 2:1 MeCN/1 M NaOH. ^bTest

strips were dipped into (i) 10 mg of RDX or (ii) 15 mg of PETN in 3 mL of MeCN or (iii) neat MeCN and irradiated at 254 nm for 1 min. ^cTest strips

were dipped into (i) 10 mg of RDX or (ii) 15 mg of PETN in 3 mL of MeCN or (iii) neat MeCN and irradiated with polychromatic light from a solar

simulator for 30 min.

2977

dx.doi.org/10.1021/jo200280c |J. Org. Chem. 2011, 76, 2976–2993

The Journal of Organic Chemistry

FEATURED ARTICLE

Scheme 2. Nitration of N,N-Dimethylaniline with the

Photofragmentation Products of RDX and PETN

Scheme 3. Route A for the Synthesis of 9,9-Disubstituted

9,10-Dihydroacridines

(to produce nitrogen dioxide (12) or nitrite, respectively) exists

and the exact nature of the initial photoreaction is ambiguous.^14,12d

Nevertheless, it can be agreed that the proposed initial products

of RDX and PETN photolysis are highly reactive, electrophilic

NO_xspecies, which can conceivably convert sulfanilamide 4 to

the diazonium cation 6 necessary to produce a positive result in

the ZellerꢀGreiss test.

Unfortunately, the Greiss test or variations thereof cannot meet

the detection requirements for RDX and PETN. First, simple

standoﬀ detection (detection at a distance) with colorimetric spot

tests is not a viable possibility because of the diﬃculty in getting a

clear optical signal returned from a purely absorptive process.

Moreover, even with optimized reagent systems, the detection

limit of the Greiss test is in the microgram regime,¹⁵which is not

competitive with existing methods to detect RDX and PETN.

Herein, we propose instead a sensing scheme that is uniquely

selective to the photolytic cleavage of nitroester and nitramine

compounds through the formation of nitroaromatic products that

provide a new ﬂuorescence signal. The pro-ﬂuorescent, or ﬂuoro-

genic, indicators described herein are capable of eﬃciently reacting

with the photofragments of RDX and PETN and constitute a new,

highly selective and sensitive detection scheme for these explosives.

was also found to produce DMNA, although longer photolysis

times (>30 min) were required and greater amounts of demethy-

lated side products were observed (most likely due to the

presence of water or other nucleophiles in the solutions).

A distinct absorbance band centered at 400 nm was found to

accompany the formation of the nitrated products under both

aerobic and anaerobic conditions, which also matched the low

energy charge-transfer band displayed by commercial DMNA.

However, DMNA has a very low ﬂuorescence quantum yield,¹⁷

and therefore, a signiﬁcant turn-on ﬂuorescence signal is not

generated upon reaction of DMA with the photofragmentation

products of RDX and PETN.

’ RESULTS AND DISCUSSION

Indicator Design. Considering the electrophilic nature of the

NO_xspecies generated by the photofragmentation of RDX and

PETN and their resemblance to the active electrophiles in

aromatic nitration reactions, we targeted reactions between

electron-rich tertiary aromatic amines and the photofragments

of RDX and PETN. It was found that photolysis (λ = 313 nm) of

a mixture of N,N-dimethylaniline (DMA) and 2 equiv of either

RDX or PETN for 10 min in acetonitrile under anaerobic

conditions afforded the formation of N,N-dimethyl-4-nitroani-

line (DMNA) in 14% yield (GC yield). Higher yields of DMNA

were obtained with longer photolysis times, and DMNA was

formed in ca. 80% yield after 1 h. The photoreaction between

DMA and either RDX or PETN under anaerobic conditions was

observed to produce only a single, yellow-colored product

(DMNA), and other side products were not evident by TLC

To probe the scope of the photonitration reaction, we investigated

whether 9,9-dioctylﬂuorene, anisole, and 1,2-dimethoxybenzene

could be nitrated by RDX and PETN. Extended photolysis (5 h)

of a mixture of 9,9-dioctylﬂuorene and either RDX or PETN in 1:1

acetonitrile:THF at either 254, 313, 334, or 356 nm failed to generate

any observable products and 9,9-dioctylﬂuorene was recovered in ca.

90% yield. Photolysis of anisole with RDX or PETN yielded only

trace amounts of 4-nitroanisole (<1% GC yield) after 4 h. Photolysis

of 1,2-dimethoxybenzene with either RDX or PETN yielded 1,2-

dimethoxy-4-nitrobenzene in only ca. 8% yield after 2 h; moreover,

this reaction did not proceed cleanly and numerous polar photo-

products were observed. Therefore, we concluded that anilines were

the best candidates for a potential indicator.

To create ﬂuorogenic indicators based on the facile nitration

reaction between aromatic amines and the photofragmentation

products of RDX and PETN, 9,9-disubstituted 9,10-dihydroa-

cridines (DHAs, Figure 2) were targeted as chemosensors. We

hypothesized that, upon nitration, the comparatively rigid DHAs

would generate donorꢀacceptor chromophores possessing high

ﬂuorescence quantum yields.¹⁸

1

or GCꢀMS analyses. The H NMR, IR, and high-resolution

mass spectra of the isolated yellow product exactly matched

those obtained for an authentic commercial sample of DMNA.

Conducting the photolysis under aerobic conditions resulted in

partial demethylation of DMA¹⁶and yielded a mixture of DMNA

and its demethylated analogue, N-methyl-4-nitroaniline (13)

(see Scheme 2). Photolysis of DMA with ammonium nitrate

Synthesis. As shown in Schemes 3ꢀ5, a series of 9,9-

disubstituted DHAs were synthesized, starting from either

2978

dx.doi.org/10.1021/jo200280c |J. Org. Chem. 2011, 76, 2976–2993

The Journal of Organic Chemistry

FEATURED ARTICLE

Scheme 4. Route B for the Synthesis of 9,9-Disubstituted

DHAs

Figure 3. X-ray crystal structure of DHA8.

(DHA10-13) that have two different substituents at the 9-posi-

tion, a spiro-DHA (DHA14), and a CF₃-containing DHA

(DHA15).

In all cases except one, adding a catalytic amount of concen-

trated sulfuric acid resulted in FriedelꢀCrafts reaction/cycliza-

tion of the respective tertiary alcohol intermediates to yield 9,9-

disubstituted DHAs. As shown in Scheme 3, we posit that this

transformation proceeds via formation of a carbocation. The

X-ray crystal structure of DHA8 thus obtained is shown in

Figure 3. The cyclization of compound 21 was uniquely challen-

ging, as the use of neither strong acids, Lewis acids, nor thionyl

chloride yielded DHA15.²¹However, it was found that reﬂuxing

a solution of 21 in POCl₃produced DHA15 in high yield.

Lastly, route C was followed to synthesize N-aryl DHAs

(Scheme 5). Copper-catalyzed N-arylation of 14 with 4-bromo-

toulene initially furnished 22, which was then reacted with 2.5

equiv of methylmagnesium bromide and catalytic concentrated

sulfuric acid. Unfortunately, the FriedelꢀCrafts cyclization of

intermediate 23 yielded a nearly statistical mixture of DHA16

and DHA17 (1:1.3 DHA16:DHA17), which could not be

N-phenylanthranilic acid methyl ester (routes A and C) or a

diphenylamine derivative¹⁹(route B). DHAs were accessed by an

acid-catalyzed cyclization of a tertiary alcohol intermediate (for

example, structure 16). In route A (Scheme 3), intermediate 16 is

accessed by a double 1,2-addition of an alkyl or aryl Grignard

reagent to either N-phenylanthranilic acid methyl ester (14) or

its N-methyl derivative (15); this strategy to synthesize DHAs

has previously been reported.²⁰In route B (Scheme 4), tertiary

alcohol intermediates 19ꢀ21 are accessed from 1,2-addition of

the aryllithium species derived from 18 to an appropriate ketone.

This strategy was adopted to synthesize unsymmetric DHAs

Scheme 5. Route C for the Synthesis of N-Aryl DHAs

2979

dx.doi.org/10.1021/jo200280c |J. Org. Chem. 2011, 76, 2976–2993

The Journal of Organic Chemistry

FEATURED ARTICLE

acceptably separated by either column chromatography or

recrystallization. Therefore, compound 24 was synthesized by

copper-catalyzed N-arylation of 14 with 2-bromomesitylene and

subsequently reacted with methylmagnesium bromide and sul-

furic acid to access DHA18.

Photophysics. The optical properties of DHA1ꢀ18 are

summarized in the Supporting Information (Table S1). The

DHAs reported herein displayed similar UVꢀvis absorption

spectra, with absorption maxima around 290 nm. Additionally,

DHA1ꢀ18 generally displayed a single emission band centered

at ca. 350 nm and were found to have similar fluorescence

quantum yields and excited-state lifetimes.

Electrochemistry. Cyclic voltammograms (CVs) of select

DHAs were recorded in CH₂Cl₂with tetrabutylammonium

hexafluorophosphate (TBAPF₆) as a supporting electrolyte and

were found to reveal behavior suggestive of irreversible chemical

transformations. The CV of DHA8 is shown in Figure 4 as a

representative example. The first anodic sweep resulted in a single

oxidation peak at ca. 1.20 V vs SCE, which can be ascribed to the

formation of the radical cation of DHA8. However, the corre-

sponding cathodic sweep revealed two cathodic peaks, arising

from the reduction of two different species in solution. Further-

more, a subsequent anodic sweep displayed two oxidation peaks.

Such behavior has been previously observed for triphenylamine

(TPA)²²and is attributed to the rapid dimerization of TPA radical

cations following oxidation; the electroactive TPA dimer thus

formed leads to the growth of an additional anodic and cathodic

peak after an initial anodic sweep. Based on assignments made for

the CV of TPA,²²the redox reactions responsible for the

individual anodic and cathodic peaks observed in the CV of

DHA8 were identified and are shown in Figure 3.

The dimerization of radical cations of DHA8 in the electro-

chemical cell to form D1 was conﬁrmed by independently synthe-

ꢀ 23

sizing D1. Oxidation of DHA8 with FeCl₃or [Et₃O^þSbCl₆

]

aﬀorded D1 in 30ꢀ40% yield (Scheme 6). This oxidation

reaction was found to selectively yield the dimeric product (by

TLC and crude ¹H NMR analyses); moreover, we were able to

recover the remaining, unreacted DHA8 upon reaction workup.

The use of hydrogen peroxide and tert-butyl-hydrogen peroxide

was also investigated; however, surprisingly, D1 was only formed

in less than 5% yield with these reagents and DHA8 was

recovered in ca. 90% yield after reaction workup. Attempts to

aﬀect an oxidative polymerization of DHA8 were not successful,

and only D1 was isolated. This observation can be explained by

the fact that D1, once formed, can be oxidized to a stable, closed-

shell dication (D1^2þ, see Figure 4) that cannot participate in

subsequent radical coupling reactions to form polymers. Dimer

D1 is a faint-yellow compound that displays an absorption band

centered at 457 nm and an emission band centered at 478 nm (Φ

0.20). The CV of D1 (see Supporting Information, Figure S1)

was found to match the second scan of the CV of DHA8 (see

Figure 4), thus conﬁrming the aforementioned assignments for

the anodic and cathodic peaks observed in the CV of DHA8.

The electrochemical behavior of DHA8 was similar to that of

the rest of the reported DHAs and also similar to the electro-

chemical behavior of DMA; i.e., the respective radical cations

dimerized in the electrochemical cell after the ﬁrst anodic sweep.

The values for the ﬁrst anodic peak potential (E_pa) and onset

Figure 4. Cyclic voltammogram of DHA8 (Pt button electrode, 0.1 M TBAPF₆in CH₂Cl₂, 100 mV/s). The redox reactions giving rise to each anodic

(AꢀC) and cathodic (D and E) peak are shown, and the ﬁrst anodic peak potential (E_pa) and onset potential (E_onset) for the ﬁrst scan are labeled.

2980

dx.doi.org/10.1021/jo200280c |J. Org. Chem. 2011, 76, 2976–2993

The Journal of Organic Chemistry

FEATURED ARTICLE

Scheme 6. Oxidative Dimerization of DHA8 To Form D1

Scheme 7. Photoreactions of Various 9,10-Dihydroacridines

with RDX and PETN^a

potential (E_onset) for the ﬁrst scan of the CVs of select DHAs,

DMA, and TPA are summarized in the Supporting Information

(Table S2). In general, similar values of E_paand E_onsetwere

observed for most DHAs; however, the electron-deﬁcient, CF₃-

containing DHA15 was an outlier and displayed signiﬁcantly

higher E_paand E_onsetvalues.

Reaction with RDX/PETN Photofragmentation Products.

The photoreactions between DHA1ꢀ18 and either RDX or

PETN were initially investigated in acetonitrile solutions. In

general, irradiating solutions containing DHA1ꢀ18 and either

RDX or PETN (which were initially colorless) at 313 nm under

aerobic conditions lead to the evolution of a bright yellow/

orange color after approximately 30 s to 5 min. Irradiating

solutions of DHA1ꢀ18 in the absence of either RDX or PETN

did not result in the same bright yellow/orange color, although

faint yellowing of the DHA solutions was noticed after greatly

extended exposure (>60 min) to UV light under aerobic

conditions.

The photolyses (λ = 313 nm) of select DHAs with a stoichio-

metric amount of either RDX or PETN were conducted on a

preparative scale in order to isolate and characterize the reaction

products formed. In these studies, long irradiation times

(generally 60 min) were employed to ensure complete reactant

conversion. TLC and GCꢀMS analyses of crude reaction

mixtures indicated that only a single, highly colored product

was formed in all cases. The yellow-orange products from the

reactions of DHA1, DHA4, and DHA18 with either RDX or

PETN were isolated by ﬂash column chromatography and

identiﬁed to be the mononitrated structures (26, 28, and 30,

respectively) shown in Scheme 7 by their ¹H NMR, FT-IR, and

high-resolution mass spectra (see the Supporting Information).

Compounds 26, 28, and 30 were isolated in 70ꢀ80% yield after

column chromatography, along with ca. 10ꢀ15% of unreacted

DHA1, DHA4, or DHA18. Similarly, DHA5 and DHA8 were

conﬁrmed to produce 27 and 29, respectively, in approximately

70% yield (GC yield) upon photolysis with RDX or PETN (30

min). Additionally, DHA1 and DHA4 were independently

nitrated under mild conditions using SiO₂/HNO₃,²⁴and the

products thus obtained were found to match those isolated from

the photoreactions of DHA1 and DHA4 with RDX/PETN.

The photoreaction of DHA2 with either RDX or PETN

yielded the nitrated product 31; however, compound 33 was

also isolated from the reaction mixture (Scheme 7). The yield of

33 was found to be somewhat dependent on the concentration of

a

The photoreduction of RDX/PETN by AcrH. has been previously

reported.⁷

DHA2, with a higher amount of 33 over 31 observed in dilute

solutions. The yield of 33 was also higher relative to that of 31

when the photolysis of DHA2 and RDX/PETN was conducted

in slightly wet acetonitrile. Compounds 31 and 33 were generally

isolated in 80% combined yield after ﬂash column chromatogra-

phy of the photoreactions between DHA2 and either RDX or

PETN. Furthermore, DHA6 was conﬁrmed to produce 32 and

34 (by GCꢀMS analysis) upon photolysis in the presence of

RDX/PETN. We hypothesize that 33 and 34 are formed as a

result of either H^•or hydride abstraction reactions between

DHA2 or DHA6 and the photodegradation products of RDX

and PETN. However, we are currently unsure as to the origins of

the concentration dependence of the yield of 33.

GCꢀMS analyses of the photoreactions between the remain-

ing DHAs (DHA3, DHA7, DHA9, DHA10ꢀ17) and either

RDX or PETN similarly revealed the formation of mononitrated

derivatives of the respective DHAs.

Other Nitroesters and Nitramines. The photoreactions

between DHA1ꢀ18 and either a model nitramine or nitroester

compound—N,N-diisopropylnitramine (NA) and amyl nitrate

(AN), respectively (Figure 5)—were also investigated. The

reaction products observed upon photolysis (λ = 313 nm) of

mixtures of DHA1-18 and either NA or AN were identical (as

established by TLC and GCꢀMS analyses) to the aforemen-

tioned nitrated products observed with RDX and PETN. How-

ever, the observed yields (GC yields) of nitrated DHAs were

significantly lower with NA/AN, as compared to RDX/PETN.

For example, whereas 26 was formed in 75% yield upon

photolysis with either RDX or PETN for 30 min, the photolysis

of DHA1 with NA or AN afforded 26 in only 30% yield under

identical reaction conditions. Therefore, it can be tentatively

inferred that RDX and PETN are more susceptible to photolytic

cleavage than their respective model compounds.

2981

dx.doi.org/10.1021/jo200280c |J. Org. Chem. 2011, 76, 2976–2993

The Journal of Organic Chemistry

FEATURED ARTICLE

Light Sources. Importantly, precise timing and sophisticated,

high-intensity light sources were not found to be necessary to

effect the reaction between DHA1-18 and the degradation

products of either RDX or PETN. Simply exposing a mixture

of DHA1-18 and RDX/PETN to polychromatic light from a

solar simulator effected the photolytic cleavage of RDX/PETN

and subsequent formation of mononitrated DHAs. For example,

compounds 26 and 28 could both be isolated in 75% yield (after

flash column chromatography) after a mixture of RDX or PETN

and DHA1 or DHA4, respectively, in acetonitrile were exposed

to simulated sunlight for 45 min. The yields of compounds 26

and 28 thus obtained are similar to those reported earlier for

photolysis at 313 nm.

Other NO_xSources. The (photo)reactions of DHA1-18 with

sodium nitrite, potassium nitrate, and NO were also investigated

to judge the limitations of using DHA1-18 as stand-off indicators

for RDX/PETN. Exposing a mixture of either DHA1, DHA5,

DHA4, or DHA8 and excess sodium nitrite in 2:1 acetonitrile/

water to simulated sunlight for 2 h did not result in significant

nitration of these DHAs (<1% GC yields of 26ꢀ29 were

generally observed). However, upon addition of 100 μL acetic

acid to the same reaction mixtures, compounds 26ꢀ29 were

formed in approximately 8% yield in the absence of light.

Protonating nitrite salts generates nitrous acid, which is known

to decompose and form HNO₃(among other species), which

most likely nitrated the DHAs in this case.

Figure 5. Structures of a model nitramine, N,N-diisopropylnitramine

(NA), and a model nitroester, amyl nitrate (AN).

Differences in DHA Reaction Mechanisms. As shown in

Scheme 7, it is interesting to note the difference in photochemical

reaction mechanisms between various 9,10-dihydroacridines. As

previously reported,⁷N-methyl-9,10-dihydroacridine (AcrH₂)

participates in a hydride transfer reaction with either RDX,

PETN, NA, or AN. Dialkylation or diarylation of the 9-position

of AcrH₂effectively nullifies its ability to donate a hydride ion

and promotes the photonitration reaction detailed herein.

Table 1. Optical Properties of Select Mononitrated DHAs

a

λ_max

a

compd

(log ε)

λ_em

Φ

DMNA 395 (3.9) 530 <0.01 (MeCN)^b0.09 (CHCl₃)^b0.17 (ﬁlm)^c,d

26

28

30

31

408 (4.1) 535

410 (4.1) 540

413 (4.2) 548

409 (4.1) 539

0.09 (MeCN)^b0.27 (CHCl₃)^b0.35 (ﬁlm)^c,d

0.10 (MeCN)^b0.30 (CHCl₃)^b0.42 (ﬁlm)^c,d

0.14 (MeCN)^b0.37 (CHCl₃)^b0.45 (ﬁlm)^c,d

0.05 (MeCN)^b0.22 (CHCl₃)^b0.33 (ﬁlm)^c,d

^aIn MeCN. ^bMeasured against perylene in EtOH (Φ 0.94). ^c10 wt %

in cellulose acetate. ^dMeasured against 10 wt % perylene in PMMA

(Φ 0.78).

The addition of a large excess of monomeric NO gas to dry,

oxygen-free solutions of the aforementioned DHAs failed to

generate the characteristic yellow-orange color of 26ꢀ29;

Figure 6. Absorption (A) and emission (B, λ_ex= 415 nm) proﬁles of the photoreaction of DHA5 with RDX in acetonitrile upon irradiation at 313 nm.

[DHA5] = 1.3 ꢁ 10^ꢀ4M. [RDX] = 5.4 ꢁ 10^ꢀ5M. Identical proﬁles are observed for the photoreactions of DHA5 with PETN. The presence or absence

of oxygen similarly does not aﬀect the observed absorption and emission proﬁles. The absorption proﬁle for the extended irradiation of a blank, aerated

solution of DHA5 is also shown (C).

2982

dx.doi.org/10.1021/jo200280c |J. Org. Chem. 2011, 76, 2976–2993

The Journal of Organic Chemistry

FEATURED ARTICLE

Figure 7. (A) Absorption proﬁle of the photoreaction of DHA5 and RDX in acetonitrile upon exposure to broad-band light from a solar simulator.

[DHA5] = 1.3 ꢁ 10^ꢀ4M. [RDX] = 5.4 ꢁ 10^ꢀ5M. (B) The rate of formation of the 408 nm absorbance peak in the presence of RDX upon exposure to

either simulated sunlight (120 mW/cm²) or monochromatic 313 nm light (10 mW/cm²).

Figure 8. Absorption (A) and emission (B, λ_ex= 470 nm) proﬁles of the photoreaction of DHA8 with RDX in acetonitrile upon exposure to simulated

sunlight. [DHA8] = 1.3 ꢁ 10^ꢀ4M. [RDX] = 5.4 ꢁ 10^ꢀ5M. The dashed green line depicts the emission spectrum obtained for a blank solution of DHA8

after irradiation under either aerobic or anaerobic conditions for 60 min. The absorption proﬁle for the irradiation of a blank, aerated solution of DHA8 is

also shown (C); the same proﬁle is also obtained for oxygen-free solutions of DHA8.

however, upon introduction of oxygen to these solutions, the

nitrated DHAs were observed to be formed by eye (in the

absence of light). Subsequent GCꢀMS analyses conﬁrmed that

26ꢀ29 were formed in ca. 20% yield. Once again, NO is known

to form nitrogen dioxide upon exposure to oxygen, which most

likely resulted in nitration of the DHAs.

Mixtures of DHA1, DHA5, DHA4, or DHA8 and a large

excess of potassium nitrate in 2:1 acetonitrile/water did not

immediately result in nitration. If allowed to stand for 24 h,

26ꢀ29, along with multiply nitrated derivatives of the afore-

mentioned DHAs, were formed in less than 5% combined yield

(GC yield). Adding acetic acid to DHA/KNO₃mixtures resulted

in the formation of multiply nitrated DHAs, with 2,7-dinitro

DHAs being the major products. Exposing mixtures of either

DHA1, DHA5, DHA4, or DHA8 and a large excess of potassium

nitrate in 2:1 acetonitrile/water to simulated sunlight for 60 min

similarly yielded multiply nitrated derivatives of these DHAs in

approximately 20% combined yield. Stoichiometric or substoi-

chiometric amounts of potassium nitrate or shorter irradiation

times failed to generate observable quantities of nitrated DHAs.

Optical Properties of Nitrated DHAs. The photophysical

properties of select nitrated DHAs, which were either isolated

from the photolysis reactions between DHAs and RDX/PETN

or synthesized by nitrating an appropriate DHA, are listed in

2983

dx.doi.org/10.1021/jo200280c |J. Org. Chem. 2011, 76, 2976–2993

The Journal of Organic Chemistry

FEATURED ARTICLE

absorption band centered at ca. 408 nm was observed to evolve

when DHA5 is photolyzed (λ 313 nm) with RDX, which

corresponds to the formation of 27. An emission band at

approximately 540 nm concomitantly evolved, which can be

assigned to emission from 27 based on the emission profile

recorded for its N-H analog 26. A ca. 27-fold increase in the

emission intensity at 540 nm was generated after 2 min of UV

irradiation. Exactly similar absorption and emission profiles were

obtained for the photoreaction between DHA5 and PETN.

Moreover, the presence or absence of oxygen did not noticeably

change the observed optical response.

Photolysis of DHA5 under aerobic conditions in the absence

of RDX/PETN failed to generate a distinct absorbance band at

408 nm. Surprisingly, electron-rich DHA5 was found to be

relatively photostable: 30 min of continuous UV irradiation did

not result in a noticeable change in the absorption spectrum of

DHA5 (Figure 6C), and its emission peak at 355 nm was found

to be bleached by only 10%. Further UV irradiation eventually

lead to slight yellowing of DHA5 solutions, and poorly deﬁned

absorbance peaks at 356 nm and ca. 440 nm appeared in the

absorption spectrum after 2 h of continuous UV irradiation

under air (Figure 6C). These new absorption peaks most likely

correspond to the formation of radical cations, N-demethylated

species, and/or N-oxide derivatives of DHA5. Notably, though, a

signiﬁcant portion of this photolyzed DHA5 solution remained

unoxidized after 2 h, and therefore, the subsequent addition of

RDX or PETN nonetheless lead to evolution of a 408 absorbance

peak and 540 nm emission peak (5-fold emission turn-on) after a

20 s exposure to 313 nm light.

As seen in Figure 7, exposing a mixture of DHA5 and RDX to

broad-band light from a solar simulator lead to the evolution of

the same 408 nm peak observed with irradiation at 313 nm. The

rate of formation of the 408 nm peak upon exposure to simulated

sunlight also matched that observed upon exposure to mono-

chromatic 313 nm light from a xenon arc lamp (Figure 7B).

Therefore, simulated sunlight was preferentially used as the light

source in subsequent studies to prove that the DHAs can

function as technology-unintensive, ﬂuorogenic indicators for

RDX/PETN under ambient sunlight.

DHA6 behaved similar to DHA5 in terms of its optical

response (the absorption and emission proﬁles for the reaction

between DHA6 and PETN upon exposure to simulated sunlight

are shown in the Supporting Information, Figure S2). Speciﬁ-

cally, an absorbance peak at 409 nm evolved in the presence of

either RDX or PETN, accompanied with evolution of an emis-

sion band at 540 nm. The presence or absence of oxygen did not

aﬀect the observed optical response to RDX/PETN. DHA6 was

also found to be relatively photostable, with no change in its

absorption spectrum and a 5% bleaching of its emission band at

382 nm observed after 30 min of continuous exposure to

sunlight. The only signiﬁcant diﬀerence between DHA5 and

DHA6 was the rate of formation of the 409 nm/540 nm

absorption/emission peaks. DHA5 was found to yield a turn-

on signal approximately three times faster than DHA6. We

hypothesize that this comparatively slow response is because

DHA6 forms a mixture of 32 and 34 upon reacting with RDX/

PETN (see Scheme 7).

Figure 9. Eﬀect of the substituents at the 9-position of DHAs on their

photoreactions with RDX and PETN. Shown are the rates of evolution

of the absorbance peak at 410 nm (470 nm for DHA8) for the

photoreactions between DHA5-15 and (A) RDX or (B) PETN. DOF

is 9,9-dioctylﬂuorene, which was used as a negative control.

Table 1. In general, the nitrated DHAs displayed similar absor-

bance bands as DMNA, with the lowest energy bands centered at

ca. 400 nm. Additionally, emission bands centered at ca. 540 nm

were observed for all isolated mononitrated DHAs. The fluores-

cence quantum yields of the compounds listed in Table 1 were

found to be solvent dependent, with the lowest quantum yields

observed in acetonitrile.²⁵Moreover, compounds 26, 28, 30, and

31 were found to display significant emission in the solid state (in

cellulose acetate films containing 10 wt % of the appropriate

compound).

The optical response of DHA18 to either RDX or PETN was

also similar to that of DHA5 (see the Supporting Information,

Figure S3). An absorbance band at 413 nm and an emission peak

at 550 nm evolved upon exposure to simulated sunlight in the

presence of either RDX or PETN. DHA18 was also relatively

Optical Characterization of Indicator Response. The ab-

sorption and emission profiles for the reaction between DHA5

and RDX under aerobic conditions are shown in Figure 6. An

2984

dx.doi.org/10.1021/jo200280c |J. Org. Chem. 2011, 76, 2976–2993

The Journal of Organic Chemistry

FEATURED ARTICLE

Figure 10. Eﬀect of the N-substituent of DHAs on their photoreactions with RDX. Shown are the rates of evolution of the absorbance peak at 410 nm

(470 nm for DHA8) for the photoreactions between various DHAs and RDX.

photostable, with no change in its absorption spectrum and a 5%

bleach of its emission band at 371 nm observed after continuous

exposure to simulated sunlight for 30 min. The rate of photo-

reaction of DHA18 with RDX/PETN was slower than that of

DHA5 but faster than that of DHA6.

of this work, it can be seen in Figure 8 that the competing

photoreaction in blank solutions of DHA8 (dashed green line) is

slower than the photonitration of DHA8 in the presence of

RDX/PETN and an emission peak at 550 nm is cleanly

generated by these explosives in under 10 s.

9,9-Diphenyl-substitutedDHA8diﬀered slightly fromthe other

DHAs explored in this work, as an absorbance band centered at

470 nm, as opposed to ca. 410 nm, evolved during its photoreac-

tion with either RDX or PETN (Figure 8). On the basis of the

accompanying GCꢀMS analyses, this absorbance band could be

assigned to the formation of 29. An emission band at 550 nm was

also observed to evolve concomitantly. An approximately 25-fold

increase in the emission intensity at 550 nm was generated in the

presence of either RDX or PETN upon exposure to simulated

sunlight for 40 s. The rates of reaction of DHA5 and DHA8 with

RDX/PETN were approximately similar.

Unlike DHA5, DHA6, and DHA18, exposing solutions of

DHA8 to sunlight (or monochromatic UV light) in either the

presence or absence of oxygen lead to the formation of a distinct

absorbance band at 380 nm, with an accompanying emission

band centered at 478 nm. The same photoreactivity was also

observed for other DHAs that contained at least one phenyl

substituent in the 9-position (DHA4, DHA11, and DHA13).

Since these absorption/emission bands were observed to evolve

even in the absence of oxygen, they are most likely not generated

by simple photooxidation products of DHA8. Moreover, the

evolution of the absorbance band at 380 nm cannot be ascribed

to a photodimerization event, as the product of such a reaction,

D1 (Scheme 6), has an absorption maximum of 457 nm. We are

currently unsure as to the origin of the photoproduct responsible

for the 380 nm/478 nm absorption/emission peak but suspect

that a photocyclization reaction occurs in DHAs with at least one

phenyl substituent in the 9-position. Nevertheless, for the purposes

Reaction Kinetics. The most significant difference between

the DHAs reported in this work involved the rate of formation of

the nitrated photoproducts upon reaction with RDX or PETN.

By following the evolution of the characteristic low-energy

charge transfer band (centered at ca. 400 nm) of the nitrated

DHAs with irradiation time, we were able to identify differences

in the reactivities of DHA1ꢀ18 (Figures 9 and 10). As can be

seen in Figure 9, the substituents at the 9-position of DHAs

significantly affected their reactivities. DHAs with at least one

methyl or phenyl substituent at the 9-position were rapidly

nitrated in the presence of RDX or PETN. DHAs with alkyl

(other than methyl) substituents at the 9-position displayed

relatively slower rates of nitration, with isopropyl substituents

leading to the slowest reaction rates. Replacing the 9-methyl

substituents with trifluoromethyl moieties also retarded the

reaction rate. Nominally faster reaction rates were generally

observed with PETN over RDX for all DHAs. 9,9-Dioctylfluor-

ene was used as a negative control for these studies, and in all

cases, the DHAs reported in this work yielded a significant

absorption signal at 400 nm over background.

The nature of the N-substituent was also found to aﬀect the

rate of photonitration in the presence of RDX/PETN. As seen in

Figure 10, for DHAs with ethyl or isopropyl substituents at the

9-position, the N-H analogues reacted faster the N-Me analo-

gues. For DHAs with phenyl or methyl substituents in the

9-position, this trend was reversed and N-Me analogues dis-

played the fastest reaction rates. Moreover, N-arylation was

found to signiﬁcantly retard the rate of photonitration.

2985

dx.doi.org/10.1021/jo200280c |J. Org. Chem. 2011, 76, 2976–2993

The Journal of Organic Chemistry

FEATURED ARTICLE

Figure 11. Comparison of the rates of nitration of DHA5 vs DMA in

the presence of (A) RDX or (B) PETN.

Lastly, the rate of formation of nitrated DHAs was compared

to the formation of DMNA from DMA. As seen in Figure 11, the

reactivity of DHA5, which displayed the fastest rate of nitration

among DHA1-18, is comparable to that of DMA.

Figure 12. (A) Emission proﬁle (λ_ex420 nm) of a glass slide coated

with DHA5 (black line) and the same slide after spotting with ca. 10 ng

of RDX and irradiating with a solar simulator for 60 s (green line).

(B) Picture of a glass slide coated with DHA5, spotted with ca. 10 ng

RDX and exposed to simulated sunlight for 120 s. (C) Limits of solid-

state detection of RDX and PETN as measured by monitoring the

change in emission intensity at 540 nm upon exposure (60 s) to

simulated sunlight. In the case of potassium nitrate, a concentrated

solution (30 mM) in acetonitrile and long exposure times (600 s) were

necessary to obtain the 8-fold increase shown.

Solid-State RDX/PETN Detection. On the basis of the pre-

viously detailed rates of nitration of DHA1-18 by the photo-

fragmentation products of RDX and PETN, we initially chose to

focus on DHA5, DHA8, DHA11, and DHA13 as potential

indicators for RDX and PETN, as they displayed the fastest rates

of reaction. Between these four DHAs, DHA5 and DHA8 were

favored because their nitrated products displayed high fluores-

cence quantum yields. We chose to use DHA5 to demonstrate

detection of RDX/PETN in the solid state; however, similar

results and detection limits were also obtained with DHA8.

In order to evaluate the utility of DHA5 as a ﬂuorescent

indicator for RDX and PETN, the solid-state response of DHA5

to RDX and PETN was investigated. For this study, glass slides

coated with DHA5 were prepared by dipcoating into 8 ꢁ 10^ꢀ3M

solutions of the indicator in acetonitrile and air drying. RDX and

PETN solutions of varying concentration were spotted onto the

surface and the slides then irradiated with a solar simulator for no

longer than 120 s.

As shown in Figure 12A, an acceptable turn-on emission signal

at 540 nm was generated by 10 ng of RDX after 60 s of irradiation

with a solar simulator. In addition to a ﬂuorescence signal, the

distinct yellow color of 27 could also be observed by eye, as

shown in Figure 12B. The limits of detection of the DHA5

chemosensor were estimated by spotting RDX or PETN solu-

tions of varying concentrations onto the DHA5-coated slides and

are shown in Figure 12C. In general, a greater emission signal at

540 nm was generated by PETN over RDX, possibly because

PETN is more susceptible to photodegradation than RDX.

Select interferents, such as ketones and aldehydes, did not

produce a signiﬁcant emission signal at 540 nm. Moreover,

consistent with observations made during the synthesis of D1,

hydrogen peroxide did not react readily with DHA5 and most

likely only formed a small quantity of the radical cation of DHA5,

which is nonemissive and therefore did not produce any emission

at 540 nm.

Aqueous potassium nitrate solutions of varying concentrations

were also spotted onto the DHA5-coated glass slides in order to

gauge the response of the DHA5 indicator to nitrate contami-

nants. Consistent with previous observations, submicromolar

solutions of potassium nitrate did not generate a signiﬁcant

emission signal at 540 nm after 1 h in either the absence of

presence of simulated solar irradiation. Using a 30 mM solution

of potassium nitrate, an approximately 8-fold increase in the

emission intensity at 540 nm was observed after a 10 min

exposure to simulated sunlight. However, given the high nitrate

concentration and relatively long irradiation time necessary to

eﬀect this emission signal, interference from nitrates during

RDX/PETN detection can, in theory, be surmounted.

2986

dx.doi.org/10.1021/jo200280c |J. Org. Chem. 2011, 76, 2976–2993

The Journal of Organic Chemistry

FEATURED ARTICLE

Within experimental error, approximately 100 pg of RDX and

PETN can be detected by the DHA5 indicator under aerobic

conditions by monitoring the emission intensity at 540 nm. In

the presence of nitrate interferents, this detection limit is con-

servatively estimated as ca. 1 ng. These detection limits, although

not low enough for the detection of equilibrium vapor, are

competitive with present transportation security systems that

make use of swipes to collect particles.

were refined anisotropically. All electrochemical measurements were

made using a quasi-internal Ag wire reference electrode submerged in

0.01 M AgNO₃/0.1 M n-Bu₄NPF₆in anhydrous MeCN. Typical CVs

were recorded using a platinum button electrode as the working

electrode and a platinum coil counter electrode. The ferrocene/ferro-

cenium (Fc/Fc^þ) redox couple was used as an external reference.

Ultravioletꢀvis absorption spectra were corrected for background signal

with either a solvent-filled cuvette (solutions) or a blank microscope

slide (films). Fluorescence spectra were measured using either right-

angle (solutions) or front-face (22.5°) detection (thin films). Fluores-

cence quantum yields were determined by the optically dilute method²⁷

using quinine sulfate in 0.1 M H₂SO₄as a standard (Φ = 0.53) and were

corrected for solvent refractive index and absorption differences at the

excitation wavelength. Fluorescence lifetimes were measured via fre-

quency modulation using a 365 nm laser diode as the light source and the

modulation of POPOP as a calibration reference. For photolysis

experiments,²⁸solutions were irradiated under air at 313 nm using

either: (1) the xenon lamp (450 W) from a fluorimeter, with the

excitation slit set to 29.4 nm (the maximum value); (2) a 500 W

Mercury Arc Lamp fitted with a 313 nm interference filter (or a 334 or

365 nm interference filter) and varying neutral density filters (0.5, 1.0, or

2.0 OD); or (3) a solar simulator equipped with a 450 W xenon arc lamp,

with a spectral output of 1.3 suns under AM 1.5 conditions. The first two

light sources were calibrated with a potassium ferric oxalate actino-

meter.²⁹For each measurement, reaction progress was also monitored in

the dark to ensure that there was no thermal contribution to the nitration

of aromatic amines by RDX and PETN. Each photolysis experiment was

performed in triplicate. N-Phenylanthranilic acid was esterified following

a literature procedure.³⁰RDX and PETN were obtained from K-9

training units, which consisted of RDX/PETN adsorbed onto sand.

RDX and PETN were extracted from the sand with spectral grade

acetonitrile and precipitated by the addition of DI water. The solids thus

isolated were recrystallized three times from chloroform/acetonitrile

and stored in the dark at ꢀ4 °C.

’ CONCLUSIONS

We have found that the nitramine-containing explosive RDX

and the nitroester-containing explosive PETN are susceptible to

photofragmentation upon exposure to sunlight to produce

reactive NO_xspecies, such as nitrogen dioxide and nitric acid.

N,N-Dimethylaniline and 9,9-disubstituted 9,10-dihydroacri-

dines (DHAs) are capable of being selectively nitrated by the

reactive, electrophilic NO_xphotofragmentation products of

RDX and PETN. This nitration reaction proceeds rapidly and

yields only one major, singly nitrated product. A roughly 25-fold

increase in the emission signal at 550 nm is observed upon

nitration of DHAs due to the generation of ﬂuorescent do-

norꢀacceptor chromophores. By monitoring the emission in-

tensity at ca. 550 nm, the presence of approximately 100 pg of

RDX or PETN can be detected within 1 min by these indicators

in the solid state upon exposure to sunlight. The photonitration

reaction presented herein is a unique and selective detection

mechanism for nitroester and nitramine explosives that is distinct

from a previously reported photoreduction reaction to detect

explosives. The rapid nitration of 9,9-diphenyl- or -dimethyl-

substituted DHA chemosensors in the presence of RDX or

PETN and the resulting strong, turn-on emission signal qualify

these DHAs as cheap, impermanent indicators for the selective,

standoﬀ identiﬁcation of nitroester and nitramine explosives.

N,N-Dimethyl-4-nitroaniline (DMNA). A mixture of 0.4 mL of

DMA and 0.1 g of either RDX or PETN were dissolved in 3.0 mL of dry,

degassed acetonitrile, and the solution was photolyzed with a xenon arc

lamp at 313 nm for 60 min. The reaction mixture was sampled every 10

min to determine the GC yield of the DMNA product. Approximately

80% of DMNA (GC yield) was formed after 60 min of photolysis. The

yellow DMNA was isolated by flash column chromatography using 50/

50 hexanes/dichloromethane as an eluent: ¹H NMR (400 MHz,

CHCl₃) δ 3.11 (s, 6H), 6.59 (d, J = 8.2 Hz, 2H), 8.09 (d, J = 8.2 Hz,

2H); ¹³C NMR (100 MHz, CHCl₃) δ 40.2, 110.2, 126.0, 136.9, 154.3;

HRMS (ESI) calcd for C₈H₁₁N₂O₂[M þ H]^þ167.0815, found

167.0819; IR (KBr plate) 695 (s), 750 (s), 820 (s), 1067 (m), 1118

(m), 1232 (m), 1347 (m), 1383 (w), 1456 (s), 1483 (s), 1582 (s), 1615

’ EXPERIMENTAL SECTION

Materials, Instrumentation, and General Experimental

Methods. Synthetic manipulations that required an inert atmosphere

(where noted) were carried out under argon using standard Schlenk

techniques. All solvents were of reagent grade or better unless otherwise

noted. All solvents used for photophysical experiments were of spectro-

scopic grade. Anhydrous tetrahydrofuran, diethyl ether, toluene, and

dichloromethane were obtained from a dry solvent system. Spectro-

scopic-grade acetonitrile was degassed and stored over 4 Å sieves. ¹H and

¹³C NMR spectra for all compounds were acquired in CHCl₃at 400 and

100 MHz, respectively. The chemical shift data are reported in units of δ

(ppm) relative to tetramethylsilane (TMS) and referenced with residual

CHCl₃. ¹⁹F NMR spectra were recorded at 380 MHz. Trichlorofluor-

omethane was used as an external standard (0 ppm), and upfield shifts

are reported as negative values. In some cases, signals associated with the

CF₃groups and proximal quaternary centers were not reported in the

¹³C NMR spectra due to CꢀF coupling and low signal-to-noise ratios.

High-resolution mass spectra (HRMS) were obtained using a peak-

matching protocol to determine the mass and error range of the

molecular ion, employing either electron impact or electrospray as the

ionization technique. GCꢀMS (electron impact mass spectrometer)

data were recorded in the temperature range of 100ꢀ350 °C under a

vacuum of at least 10^ꢀ5Torr. GC retention times are reported in

minutes. X-ray crystal structures were determined with graphite-mono-

chromated Mo KR radiation (λ = 0.71073 Å). All structures were solved

by direct methods using SHELXS²⁶and refined against F on all data by

full-matrix least-squares with SHELXL-97. All non-hydrogen atoms

(w), 2924 (m) cm^ꢀ1

.

General Procedure for the Synthesis of 9,9-Disubstituted

9,10-Dihydroacridines (DHA1-4). A flame-dried Schlenk flask was

charged with 1.0 g of methyl N-phenylanthranilate (14, 4.4 mmol) and

45 mL of dry, degassed Et₂O under argon and cooled to 0 °C in an ice

bath. A 3.5 equiv portion of the appropriate Grignard reagent in Et₂O

was added dropwise and the reaction mixture allowed to stir at room

temperature under argon for 3 d. After the mixture was quenched with

saturated ammonium chloride, the organic layer was separated, washed

with brine and water, and dried over MgSO₄and the solvent evaporated

under reduced pressure. The crude tertiary alcohol thus formed was

carried on to the next step without purification. To the neat oil isolated

from the previous step was added 1ꢀ2 mL of concentrated H₂SO₄

under argon and the reaction stirred at room temperature for 1 h under

argon. After dilution with 30 mL of DI H₂O, the reaction was poured

into a 10% (v/v) aqueous ammoniacal solution and extracted with ether

(5 ꢁ 50 mL). The combined organic layers were washed with saturated

2987

dx.doi.org/10.1021/jo200280c |J. Org. Chem. 2011, 76, 2976–2993

The Journal of Organic Chemistry

FEATURED ARTICLE

sodium bicarbonate, brine, and water and dried over MgSO₄, and the

solvent was evaporated under reduced pressure. The residue was purified

by flash column chromatography to yield the desired compound.

of a yellow oil: ¹H NMR (400 MHz, CHCl₃) δ 3.28 (s, 3H), 3.58 (s,

3H), 6.63 (dd, J = 8.8, 1.2 Hz, 2H), 6.73 (td, J = 6, 1.2 Hz, 1H), 7.16 (td,

J = 7.2, 1.6 Hz, 2H), 7.27 (m, 2H), 7.53 (td, J = 8, 1.6 Hz, 1H), 7.79 (dd,

J = 7.6, 1.6 Hz, 1H); ¹³C NMR (100 MHz, CHCl₃) δ 40.5, 52.3, 114.4,

118.1, 125.4, 129.1, 129.2, 129.4, 131.6, 133.4, 148.3, 129.4, 167.7;

HRMS (ESI) calcd for C₁₅H₁₅NO₂[M þ H]^þ242.1176, found

242.1170; IR (KBr plate) 2924 (s), 2853 (s), 1728 (m), 1594 (m),

1500 (m), 1454 (m), 1391 (w), 1253 (s), 1214 (s), 1062 (m), 1005 (m),

9,9-Dimethyl-9,10-dihydroacridine (DHA1). Synthesized

using 3.0 M methylmagnesium bromide in Et₂O and purified by flash

column chromatography using gradient elution, starting with 10%

dichloromethane in hexanes and progressing to 50% dichloromethane

in hexanes: 0.41 g (45%) of a white solid was isolated; mp 120 °C; ¹H

NMR (400 MHz, CHCl₃) δ 1.54 (s, 6H), 6.11 (s, 1H), 6.67 (dd, J = 0.8

Hz, 7.6 Hz, 2H), 6.90 (m, 2H), 7.09 (m, 2H), 7.37 (d, J = 7.6 Hz, 2H);

¹³C NMR (100 MHz, CHCl₃) δ 30.7, 36.4, 113.6, 120.8, 125.7, 126.9,

129.3, 138.6. HRMS (ESI) calc for C₁₅H₁₅N [MþH]^þ210.1277, found

210.1284. IR(KBr plate) 745 (s), 886 (m), 1037 (m), 1318 (m), 1452

(m), 1479 (s), 1507 (m), 1580 (m), 1606 (m), 2966 (m), 3359

758 (m), 575 (m) cm^ꢀ1

.

General Procedure for the Synthesis of 9,9-Disubstituted

10-Methyl-9,10-dihydroacridines (DHA5ꢀ9). A flame-dried

Schlenk flask was charged with 1.0 g of methyl N-methyl-N-phenylan-

thranilate (15, 4.1 mmol) and 45 mL of dry, degassed Et₂O under argon

and cooled to 0 °C in an ice bath. A 2.5 equiv portion of the appropriate

Grignard reagent in Et₂O was added dropwise and the reaction allowed

to stir at room temperature under argon for 3 d. After the reaction was

quenched with saturated ammonium chloride, the organic layer was

separated, washed with brine and water, and dried over MgSO₄and the

solvent evaporated under reduced pressure. The crude tertiary alcohol

thus formed was carried on to the next step without purification. To the

neat oil isolated from the previous step was added 1ꢀ2 mL of

concentrated H₂SO₄under argon and the reaction stirred at room

temperature for 1 h under argon. After dilution with 30 mL of DI H₂O

the reaction was poured into a 10% (v/v) aqueous ammoniacal solution

and extracted with ether (5 ꢁ 50 mL). The combined organic layers

were washed with saturated sodium bicarbonate, brine, and water and

dried over MgSO₄, and the solvent was evaporated under reduced

pressure. The residue was purified by flash column chromatography to

yield the desired compound.

(m) cm^ꢀ1

.

9,9-Diethyl-9,10-dihydroacridine (DHA2). Synthesized using

3.0 M ethylmagnesium bromide in Et₂O and purified by flash column

chromatography using hexanes as the eluent: 0.36 g (35%) of a clear oil

was isolated; ¹H NMR (400 MHz, CHCl₃) δ 0.91 (t, J = 7.6 Hz, 3H),

0.97 (t, J = 7.6 Hz, 3H), 2.24 (q, J = 7.6 Hz, 2H), 2.35 (q, J = 7.6 Hz, 2H),

5.74 (s, 1H), 6.90 (m, 3H), 7.06 (m, 5H); ¹³C NMR (100 MHz, CHCl₃)

δ 13.0, 13.1, 13.9, 14.8, 24.3, 31.7, 115.9, 116.8, 118.2, 118.5, 120.3,

120.5, 121.0, 121.2, 123.0, 124.9, 127.4, 127.5, 129.5, 129.7, 130.1, 130.2,

134.3, 140.1, 140.3, 141.2, 143.5, 143.7; HRMS (ESI) calc for C₁₇H₁₉N

[M þ H]^þ238.1590, found 238.1591; IR (KBr plate) 692 (m), 745 (s),

1309 (m), 1451 (m), 1507 (s), 1575 (m), 1593 (s), 2871 (m), 2930 (m),

2965 (m), 3041 (m), 3403 (m) cm^ꢀ1

.

9,9-Diisopropyl-9,10-dihydroacridine (DHA3). Synthesized

using 2.0 M isopropylmagnesium chloride in Et₂O and purified by flash

column chromatography using hexanes as the eluent: 0.19 g (17%) of a

clear oil was isolated; ¹H NMR (400 MHz, CHCl₃) δ 0.77 (d, J = 6.8 Hz,

3H), 1.02 (d, J = 6.8 Hz, 3H), 1.46 (s, 3H), 1.87 (s, 3H), 3.06 (septet, J =

6.8 Hz, 1H), 5.75 (s, 1H), 6.88 (m, 5H), 7.20 (m, 4H); ¹³C NMR (100

MHz, CHCl₃) δ 19.7, 20.9, 22.4, 22.2, 30.6, 115.5, 118.6, 119.8, 121.2,

127.1, 129.5, 129.8, 130.0, 130.7, 136.8, 141.0, 143.5; HRMS (ESI) calc

for C₁₉H₂₄N [M þ H]^þ266.1903, found 266.1904; IR (KBr plate) 693

(s), 745 (s), 1079 (m), 1309 (s), 1450 (s), 1508 (s), 1576 (s), 1594 (s),

9,9-Dimethyl-10-methyl-9,10-dihydroacridine (DHA5).

Synthesized using 3.0 M methylmagnesium bromide in Et₂O and

purified by flash column chromatography using gradient elution, starting

with 10% dichloromethane in hexanes and progressing to 50% dichlor-

omethane in hexanes: 0.39 g (42%) of a light yellow solid was isolated;

mp 93 °C; ¹H NMR (400 MHz, CHCl₃) δ 1.52 (s, 6H), 3.43 (s, 3H),

6.96 (m, 4H), 7.21 (m, 2H), 7.38 (d, J = 1.6 Hz, 2H); ¹³C NMR (100

MHz, CHCl₃) δ 27.4, 33.5, 36.7, 112.3, 120.8, 123.8, 126.8, 132.8, 142.4;

HRMS (ESI) calc for C₁₆H₁₇N [M þ H]^þ224.1434, found 224.1429;

IR (KBr plate) 751 (s), 1046 (m), 1268 (s), 1340 (s), 1450 (s), 1470 (s),

2961 (m), 3399 (s) cm^ꢀ1

.

9,9-Diphenyl-9,10-dihydroacridine (DHA4). Synthesized

using 3.0 M phenylmagnesium bromide in Et₂O and purified by flash

column chromatography using gradient elution, starting with 10%

dichloromethane in hexanes and progressing to 50% dichloromethane

in hexanes: 0.83 g (57%) of a white solid was isolated; mp 230 °C; ¹H

NMR (400 MHz, CHCl₃) δ 6.25 (s, 1H), 6.86 (m, 9H), 7.16 (m, 8H);

¹³C NMR (100 MHz, CHCl₃) δ 30.1, 56.7, 113.5, 120.2, 125.6, 126.1,

126.2, 127.1, 127.4, 127.6, 127.6, 127.9, 128.5, 130.0, 130.2, 130.2, 139.7,

146.0, 149.3; HRMS (ESI) calc for C₂₅H₁₉N [M þ H]^þ334.1590,

found 334.1584; IR (KBr plate) 699 (m), 734 (m), 753 (m), 907 (m),

1590 (s), 2900 (m), 2950 (s), 2980 (s), 3050 (m) cm^ꢀ1

.

9,9-Diethyl-10-methyl-9,10-dihydroacridine (DHA6).

Synthesized using 3.0 M ethylmagnesium bromide in Et₂O and purified

by flash column chromatography using hexanes as the eluent. 0.31 g

(30%) of a clear oil was isolated; ¹H NMR (400 MHz, CHCl₃) δ 0.81 (t,

J = 7.6 Hz, 3H), 0.84 (t, J = 7.6 Hz, 3H), 2.14 (q, J = 7.6 Hz, 2H), 2.27 (q,

J = 7.6 Hz, 2H), 3.06 (s, 3H), 6.65 (m, 3H), 7.17 (m, 5H); ¹³C NMR

(100 MHz, CHCl₃) δ 14.2 (2), 14.3, 22.5, 22.8, 23.2, 23.4, 29.6, 30.4,

31.0, 31.2, 31.7, 37.1, 39.4, 39.6, 114.0, 114.1, 117.2, 117.3, 125.5, 125.9,

127.8, 128.1, 128.2, 128.4, 128.7, 128.9, 130.5, 132.0, 132.5, 140.0, 140.1,

141.5, 143.4, 146.1, 147.0, 149.3, 149.5; HRMS (ESI) calc for C₁₈H₂₁N

[M þ H]^þ252.1747, found 252.1742; IR (KBr plate) 692 (s), (748 (s),

1342 (m), 1444 (m), 1487 (s), 1500 (s), 1568 (m), 1592 (s), 1602 (s),

1315 (m), 1474 (s), 1604 (m), 3057 (w), 3393 (m) cm^ꢀ1

.

Methyl N-Methyl-N-phenylanthranilate (15). A flame-dried

two-neck round-bottom flask was charged with 8 g of N-phenylanthra-

nilic acid (37.5 mmol), 0.3 mL of 15-crown-5, 300 mL of dry THF, and

100 mL of dimethoxyethane under argon. The solution was cooled to

0 °C in an ice bath, 5 g of a 60 wt % dispersion of NaH in mineral oil (3 g

NaH, 125 mmol) was added to the reaction mixture in small portions

under argon, and 15 mL dimethyl sulfate (19.99 g, 158 mmol) was added

via syringe. After being stirred at room temperature for 5 d under argon,

the reaction mixture was poured carefully onto 800 g of ice and extracted

with Et₂O (5 ꢁ 50 mL). The organic layers were combined, washed

thoroughly with saturated sodium bicarbonate (3 ꢁ 25 mL), brine, and

water, and dried over MgSO₄, and the solvent was evaporated under

reduced pressure. The resulting oil was purified by flash column

chromatography using gradient elution, starting with 100% hexanes

and progressing to 30% dichloromethane in hexanes to yield 7.2 g (80%)

2810 (m), 2870 (m), 2963 (m), 3024 (m) cm^ꢀ1

.

9,9-Diisopropyl-10-methyl-9,10-dihydroacridine (DHA7).

Synthesized using 2.0 M isopropylmagnesium chloride in Et₂O and

purified by flash column chromatography using hexanes as the eluent:

0.19 g (17%) of a clear oil was isolated; ¹H NMR (400 MHz, CHCl₃) δ

0.85 (d, J = 7.2 Hz, 3H), 0.89 (d, J = 7.6 Hz, 3H), 1.46 (s, 3H), 1.80 (s,

3H), 2.84 (septet, J = 7.2 Hz, 1H), 3.03 (s, 3H), 6.74 (m, 3H), 6.98 (m,

1H), 7.13 (m, 5H); ¹³C NMR (100 MHz, CHCl₃) δ 21.6, 22.3, 23.8,

31.8, 39.0, 114.6, 117.4, 120.6, 121.4, 125.2, 127.4, 127.8, 128.0, 128.7,

129.4, 133.6, 140.0, 141.2, 147.2, 149.6; HRMS (ESI) calc for C₂₀H₂₅N

[M þ H]^þ280.2060, found 280.2058; IR (KBr plate) 748 (m), 1499

(m), 1601 (m), 2820 (m), 2910 (m) 3040 (m) cm^ꢀ1

.

2988

dx.doi.org/10.1021/jo200280c |J. Org. Chem. 2011, 76, 2976–2993

The Journal of Organic Chemistry

FEATURED ARTICLE

9,9-Diphenyl-10-methyl-9,10-dihydroacridine (DHA8).

Synthesized using 3.0 M phenylmagnesium bromide in Et₂O and

purified by flash column chromatography using gradient elution, starting

with 10% dichloromethane in hexanes and progressing to 50% dichlor-

omethane in hexanes: 0.79 g (55%) of a light yellow solid was isolated;

mp 165ꢀ166 °C; ¹H NMR (400 MHz, CHCl₃) δ 3.29 (s, 3H), 6.84 (m,

2H), 6.91 (m, 8H), 7.18 (m, 6H), 7.26 (m, 2H); ¹³C NMR (100 MHz,

CHCl₃) δ 33.6, 57.3, 112.1, 120.0, 126.4, 127.4, 127.7, 130.1, 130.6,

131.4, 142.7, 146.2; HRMS (ESI) calc for C₂₆H₂₁N [M þ H]^þ

348.1747, found 348.1732; IR (KBr plate) 638 (m), 699 (m), 733

(m), 755 (m), 1270 (m), 1348 (m), 1468 (s), 1590 (m), 1589 (m), 2815

temperature for 1 h under argon. After dilution with 30 mL of DI H₂O,

the reaction was extracted with Et₂O (5 ꢁ 50 mL), the organic layers

were combined, washed with brine and water and dried over anhydrous

MgSO₄, and the solvent was evaporated under reduced pressure. The

residue was purified by flash column chromatography using hexanes as

the eluent to yield 0.57 g (57%) of the desired compound as a clear oil:

¹H NMR (400 MHz, CHCl₃) δ 1.66 (dd, J = 6.8, 0.8 Hz, 3H), 1.78 (t, J =

1.2 Hz, 3H), 3.10 (s, 3H), 5.49 (m, 1H), 6.66 (m, 3H), 7.18 (m, 6H);

¹³C NMR (100 MHz, CHCl₃) δ 14.2, 15.2, 16.5, 24.2, 39.2, 39.6, 113.8,

114.1, 117.1, 117.4, 122.4, 124.4, 125.7, 125.9, 128.0, 128.2, 128.5, 128.8,

128.9, 130.6, 131.5, 136.4, 140.4, 144.3, 145.8, 146.9, 149.2, 149.4;

HRMS (ESI) calc for C₁₇H₁₉N [M þ H]^þ238.1590, found 238.1587;

IR (KBr plate) 692 (m), 747 (m), 1345 (m), 1444 (m), 1499 (s), 1592

(w), 2873 (w), 3056 (m) cm^ꢀ1

.

9,9-Di-n-octyl-10-methyl-9,10-dihydroacridine (DHA9).

Synthesized using 2.0 M octylmagnesium bromide in Et₂O and purified

by flash column chromatography using hexanes as the eluent: 0.7 g

(40%) of a clear oil was isolated; ¹H NMR (400 MHz, CHCl₃) δ 0.85 (t,

J = 7.6 Hz, 3H), 0.88 (t, J = 7.2 Hz, 3H), 1.22 (bm, 22H), 2.08 (q, J = 7.2

Hz, 2H), 2.25 (t, J = 7.6 Hz, 2H), two overlapping singlets: δ 3.08, 3.10,

total 3H, 5.35 (t, J = 7.6 Hz, 1H), 6.67 (m, 3H), 7.17 (m, 6H); ¹³C NMR

(100 MHz, CHCl₃) δ 14.4, 22.9 (3), 28.3, 28.7, 29.0, 29.3, 29.5 (2), 29.6,

29.7 (2), 29.8, 30.0, 30.3, 31.1, 32.1 (2), 37.4, 39.4, 39.6, 114.0, 114.1,

117.2, 117.3, 125.5, 125.9, 128.0 (2), 128.2, 128.4, 128.6, 128.9,130.7,

132.0, 132.5, 140.0, 140.7, 141.3, 143.4, 146.1,146.9, 149.3, 149.5;

HRMS (ESI) calc for C₃₀H₄₅N [M þ H]^þ420.3625, found

420.3613; IR (KBr plate) 692 (m), 747 (m), 1095 (w), 1342 (m),

1499 (s), 1592 (s), 1602 (s), 2854 (s), 2924 (s), 2955 (s), 3024

(s), 1602 (s), 2918 (m), 3024 (m) cm^ꢀ1

.

9-Methyl-9-phenyl-10-methyl-9,10-dihydroacridine (DHA11).

A flame-dried two-necked flask was charged with 2-bromo-N-methyl-N-

phenylaniline (1.0 g, 3.8 mmol) and 100 mL of dry THF under argon

and cooled to ꢀ78 °C in a dry iceꢀacetone bath. A 2.6 mL portion of a

1.6 M solution of n-BuLi in hexanes (4.16 mmol) was added dropwise

over 5 min and the reaction stirred at ꢀ78 °C under argon for 1 h.

Acetophenone (0.49 mL, 0.51 g, 4.2 mmol) was added in one portion

under argon at ꢀ78 °C, and the reaction allowed to warm to room

temperature and stirred overnight under argon. After being quenched with

saturated ammonium chloride, the reaction mixture was extracted with

Et₂O, the organic layers were combined, washed with brine and water,

and dried over anhydrous MgSO₄, and the solvent evaporated under

reduced pressure. To the neat residue thus obtained was added 2 mL of

concentrated H₂SO₄under argon and the mixture stirred at room

temperature for 1 h under argon. After dilution with 30 mL of DI H₂O,

the reaction mixture was extracted with Et₂O (5 ꢁ 50 mL), the organic

layers were combined, washed with brine, and water, and dried over

anhydrous MgSO₄, and the solvent was evaporated under reduced

pressure. The desired product was purified by flash column chromatog-

raphy using gradient elution, starting with 20% dichloromethane in

hexanes and progressing to 50% dichloromethane in hexanes to yield

0.55 g (50%) of a light yellow solid: mp 184 °C; ¹H NMR (400 MHz,

(w) cm^ꢀ1

.

2-Bromo-N-methyl-N-phenylaniline (18). A flame-dried two-

neck round-bottom flask was charged with 2-bromo-N-phenylaniline

(3 g, 12.2 mmol), 0.1 mL of 15-crown-5, 200 mL of dry THF, and 50 mL

of dimethoxyethane under argon. The solution was cooled to 0 °C in an

ice bath, 0.6 g of a 60 wt % dispersion of NaH in mineral oil (0.36 g NaH,

15 mmol) was added to the reaction mixture in small portions under

argon, and 1.4 mL of dimethyl sulfate (1.89 g, 15 mmol) was added via

syringe. After being refluxed for 20 h under argon, the reaction mixture

was poured carefully onto 500 g of ice and extracted with Et₂O (5 ꢁ

50 mL). The organic layers were combined, washed thoroughly with

saturated sodium bicarbonate (3 ꢁ 25 mL), brine, and water, and dried

over MgSO₄, and the solvent was evaporated under reduced pressure.

The resulting oil was purified by flash column chromatography using

gradient elution, starting with 100% hexanes and progressing to 30%

dichloromethane in hexanes to yield 2.5 g (80%) of a clear oil: ¹H NMR

(400 MHz, CHCl₃) δ 3.22 (s, 3H), 6.56 (d, J = 7.6 Hz, 2H), 6.75 (t, J =

7.6 Hz, 1H), 7.15 (m, 3H), 7.25 (dd, J = 8.0, 2.0 Hz, 1H), 7.32 (td, J = 7.6,

1.2 Hz, 1H), 7.66 (dd, J = 8.0, 1.6 Hz, 1H); ¹³C NMR (100 MHz,

CHCl₃) δ 39.1, 113.5, 117.8, 124.4, 127.9, 128.4, 129.1, 129.2, 130.6,

134.3, 147.0, 148.7; HRMS (ESI) calcd for C₁₃H₁₂BrN [M þ H]^þ

262.0226, found 262.0234; IR (KBr plate) 654 (s), 691 (s), 748 (s), 872

(s), 1139 (s), 1137 (s), 1273 (s), 1346 (s), 1438 (m), 1467 (s), 1499 (s),

CHCl₃) δ 1.83 (s, 3H), 3.37 (s, 3H), 6.93 (m, 6H), 7.24 (m, 8H); ¹³

C

NMR (100 MHz, CHCl₃) δ 27.4, 28.2, 31.2, 33.6, 112.1 (2), 112.3,

113.7, 113.9, 116.9, 120.0, 120.4, 120.8, 123.8, 126.2, 126.4, 126.76,

126.7, 127.1, 127.38, 127.5, 127.7, 127.8, 127.9, 128.0, 128.5, 128.7,

128.8, 128.9, 130.1, 130.5, 131.4, 132.5, 142.0, 142.7, 146.2, 148.9;

HRMS (ESI) calc for C₂₁H₁₉N [M þ H]^þ286.1590, found 286.1590;

IR (KBr plate) 650 (m), 699 (m), 743 (m), 798 (m), 1290 (m), 1353

(m), 1468 (s), 1595 (m), 1631 (m), 2803 (w), 2898 (w), 3042

(m) cm^ꢀ1

.

9-Isopropyl-9,10-dimethyl-9,10-dihydroacridine (DHA12).

A flame-dried two-necked flask was charged with 2-bromo-N-methyl-

N-phenylaniline (1.0 g, 3.8 mmol) and 100 mL of dry THF under argon

and cooled to ꢀ78 °C in a dry iceꢀacetone bath. A 2.6 mL portion of a

1.6 M solution of n-BuLi in hexanes (4.16 mmol) was added dropwise

over 5 min and the reaction stirred at ꢀ78 °C under argon for 1 h.

Isopropyl methyl ketone (0.361 g, 4.2 mmol) was added in one portion

to the reaction mixture at ꢀ78 °C and the reaction allowed to warm to

room temperature and stirred overnight under argon. After being

quenched with saturated ammonium chloride, the reaction mixture

was extracted with Et₂O, the organic layers werecombined, washed with

brine and water, and dried over anhydrous MgSO₄, and the solvent was

evaporated under reduced pressure. To the neat residue thus obtained

was added 2 mL of concentrated H₂SO₄under argon and the mixture

stirred at room temperature for 1 h under argon. After dilution with

30 mL of DI H₂O, the reaction was extracted with Et₂O (5 ꢁ 50 mL),

the organic layers were combined, washed with brine, and water, and

dried over anhydrous MgSO₄, and the solvent was evaporated under

reduced pressure. The residue was purified by flash column chromatography

1580 (s), 1601 (s), 2813 (m), 2824 (s), 3058 (m), 3089 (m) cm^ꢀ1

.

9-Ethyl-9,10-dimethyl-9,10-dihydroacridine (DHA10). A

flame-dried two-necked flask was charged with 2-bromo-N-methyl-N-

phenylaniline (1.0 g, 3.8 mmol) and 100 mL dry THF under argon and

cooled to ꢀ78 °C in a dry iceꢀacetone bath. A 2.6 mL portion of a 1.6 M

solution of n-BuLi in hexanes (4.16 mmol) was added dropwise over 5

min, and the reaction mixture stirred at ꢀ78 °C under argon for 1 h.

Methyl ethyl ketone (0.302 g, 4.2 mmol) was added in one portion to the

reaction mixture at ꢀ78 °C, and the reaction allowed to warm to room

temperature and stirred overnight under argon. After being quenched

with saturated ammonium chloride, the reaction was extracted with

Et₂O, the organic layers were combined, washed with brine and water,

and dried over anhydrous MgSO₄, and the solvent was evaporated under

reduced pressure. To the neat residue thus obtained was added 2 mL of

concentrated H₂SO₄under argon and the mixture stirred at room

2989

dx.doi.org/10.1021/jo200280c |J. Org. Chem. 2011, 76, 2976–2993

The Journal of Organic Chemistry

FEATURED ARTICLE

using hexanes as the eluent to yield 0.62 g (60%) of the desired

compound as a clear oil: H NMR (400 MHz, CHCl₃) δ 1.51 (s,

IR (KBr plate) 693 (m), 747(m), 1069 (s), 1155 (m), 1263 (m), 1345

1

(m), 1499 (s), 1602(s), 2853 (m), 2921 (s) cm^ꢀ1

.

3H), 1.64 (s, 3H), 1.70 (s, 3H), 3.04 (s, 3H), 6.64 (m, 3H), 7.12 (m,

6H); ¹³C NMR (100 MHz, CHCl₃) δ 21.5, 22.2, 23.8, 31.8, 39.0, 114.6,

117.4, 120.6, 121.4, 125.1, 127.4, 127.8, 128.0, 128.7, 129.3, 133.6, 140.0,

141.1, 147.2, 149.5; HRMS (ESI) calc for C₁₈H₂₁N [MþH]^þ252.1747,

found 252.1742; IR (KBr plate) 692 (m), 747 (s), 1137 (m), 1342 (m),

1443 (m), 1486 (s), 1499 (s), 1591 (s), 1601 (s), 2911 (m), 2984

1,1,1,3,3,3-Hexafluoro-2-(2-(methyl(phenyl)amino)phenyl)-

propan-2-ol (21). A flame-dried two-necked flask equipped with a dry

ice/acetone condenser was charged with 2-bromo-N-methyl-N-pheny-

laniline (1.0 g, 3.8 mmol) and 100 mL dry THF under argon and cooled

to ꢀ78 °C in a dry iceꢀ-acetone bath. A 2.6 mL portion of a 1.6 M

solution of n-BuLi in hexanes (4.16 mmol) was added dropwise over 5

min and the reaction stirred at ꢀ78 °C under argon for 1 h. Making sure

that the dry ice/acetone condenser remained filled, anhydrous hexa-

fluoroacetone (HFA) gas was bubbled into the reaction flask at ꢀ78 °C

under a positive pressure of argon for a total duration of 3 min; the

pressure reading on the HFA tank was noted to be 22 psi. The reaction

was allowed to warm to room temperature and the HFA allowed to

reflux for an additional 3 h (making sure the dry ice/acetone condenser

remained full for the duration) after which the excess HFA was removed

by bubbling through a saturated KOH solution for 1 h. The reaction was

quenched with saturated ammonium chloride and extracted with Et₂O

(3 ꢁ 50 mL). The organic layers were combined, washed with brine and

water, and dried over MgSO₄, and the solvent was evaporated under

reduced pressure. The residue was purified by flash column chromatog-

raphy using 50% dichloromethane in hexanes as the eluent to yield g 1.1

g (80%) of the desired compound as a white crystalline solid after drying

in vacuo for 3d: ¹H NMR (400 MHz, CHCl₃) δ 3.08 (s, 3H), 6.87 (dd,

J = 8.0 Hz, J = 1.6 Hz, 2H), 6.98 (m, 2H), 7.21 (m, 2H), 7.36 (m, 2H),

7.75 (d, J = 8.0 Hz, 1H), 11.51 (s, 1H); ¹³C NMR (100 MHz, CHCl₃) δ

40.3, 80.3 (quintet), 115.0, 115.5, 118.8, 119.2 (2), 119.4, 120.6 121.5,

121.6, 122.0, 122.1, 122.9, 124.9 (3), 127.2, 127.7, 127.8, 127.9, 128.3,

128.9 (2), 129.0, 129.2, 129.3, 129.4, 129.6, 132.3, 149.2, 151.6. ¹⁹F

NMR (380 MHz, CHCl₃) δ ꢀ76.4, ꢀ75.1; HRMS (ESI) calc for

C₁₆H₁₃F₆NO [M þ H]^þ350.0974, found 350.0961; IR (KBr plate) 479

(m), 693 (s), 709 (s), 754 (s), 848 (m), 936 (m), 954 (s), 968 (s), 1121

(s), 1147 (m), 1192 (s), 1260 (s), 1496 (s), 1577 (m), 1603 (m), 2719

(m) cm^ꢀ1

.

9-Ethyl-9-phenyl-10-methyl-9,10-dihydroacridine (DHA13).

A flame-dried two-necked flask was charged with 2-bromo-N-methyl-N-

phenylaniline (1.0 g, 3.8 mmol) and 100 mL of dry THF under argon

and cooled to ꢀ78 °C in a dry iceꢀacetone bath. A 2.6 mL portion of a

1.6 M solution of n-BuLi in hexanes (4.16 mmol) was added dropwise

over 5 min and the reaction mixture stirred at ꢀ78 °C under argon for 1

h. Propiophenone (0.56 mL, 0.56 g, 4.2 mmol) was added in one portion

to the reaction mixture under argon at ꢀ78 °C and the reaction allowed

to warm to room temperature and stirred overnight under argon. After

quenching with saturated ammonium chloride, the reaction was ex-

tracted with Et₂O, the organic layers were combined, washed with brine

and water, and dried over anhydrous MgSO₄, and the solvent was

evaporated under reduced pressure. To the neat residue thus obtained

was added 2 mL of concentrated H₂SO₄under argon and the mixture

stirred at room temperature for 1 h under argon. After dilution with

30 mL of DI H₂O, the reaction was extracted with Et₂O (5 ꢁ 50 mL),

the organic layers were combined, washed with brine and water, and

dried over anhydrous MgSO₄, and the solvent was evaporated under

reduced pressure. The desired product was purified by flash column

chromatography using 10% dichloromethane in hexanes as the eluent to

yield 0.62 g (55%) of a light yellow oil: ¹H NMR (400 MHz, CHCl₃) δ

0.74 (t, J = 7.2 Hz, 3H), 2.22 (q, J = 7.2 Hz, 2H), 3.38 (s, 3H), 6.75 (m,

4H), 6.89 (m, 2H), 7.17 (m, 3H), 7.27 (m, 4H); ¹³C NMR (100 MHz,

CHCl₃) δ 9.7, 33.8, 34.6, 50.5, 111.9, 113.1, 113.8, 114.0, 116.9, 119.9,

125.4, 125.6, 125.9, 126.4, 126.5, 126.6, 126.7, 127.0, 127.3, 127.5, 127.8,

127.9, 128.2, 128.3, 128.8, 128.9, 129.2, 129.4, 129.6, 130.2, 132.9, 138.7,

141.3, 142.0, 143.5, 147.3, 148.5, 149.5; HRMS (EI) calcd for C₂₂H₂₁N

[M]^þ299.1669, found 299.1673; IR (KBr plate) 633 (m), 693 (s), 747

(s), 1032 (w), 1260 (w), 1349 (m), 1448 (m), 1499 (s), 1575 (m), 1590

(m), 2973 (m), 3066 (m), 3854 (broad) cm^ꢀ1

.

10-Methyl-9,9-bis(trifluoromethyl)-9,10-dihydroacridine

(DHA15). 1,1,1,3,3,3-Hexafluoro-2-(2-(methyl(phenyl)amino)phenyl)-

propan-2-ol (0.2 g, 0.57 mmol) was dissolved in 15 mL of POCl₃and the

solution refluxed under argon for 3 d. Excess POCl₃was distilled off

using a short path distillation head, the residue was dissolved in CHCl₃

and poured into a 10% (v/v) aqueous ammoniacal solution, and the

biphasic system was stirred at room temperature for 1 h. The organic

layer was separated and the aqueous layer extracted with Et₂O (3 ꢁ

50 mL). The organic layers were combined, washed with brine and

water, and dried over MgSO₄, and the solvent was evaporated under

reduced pressure. The residue was purified by flash column chromatog-

raphy using 10% dichloromethane in hexanes as the eluent to yield 0.15 g

(80%) of the desired compound as a light-blue oil: ¹H NMR (400 MHz,

(s), 1601 (s), 2927 (m), 3024 (m) cm^ꢀ1

.

10-Methyl-10H-spiro[acridine-9,1⁰-cyclohexane] (DHA14). A

flame-dried two-necked flask was charged with 2-bromo-N-methyl-N-

phenylaniline (1.0 g, 3.8 mmol) and 100 mL of dry THF under argon

and cooled to ꢀ78 °C in a dry iceꢀacetone bath. A 2.6 mL portion of a

1.6 M solution of n-BuLi in hexanes (4.16 mmol) was added dropwise

over 5 min and the reaction stirred at ꢀ78 °C under argon for 1 h.

Cyclohexanone (0.43 mL, 0.412 g, 4.2 mmol) was added in one portion

to the reaction mixture at ꢀ78 °C and the reaction allowed to warm to

room temperature and stirred overnight under argon. After being

quenched with saturated ammonium chloride, the reaction mixture

was extracted with Et₂O, the organic layers were combined, washed with