Addition of dimethyl phosphite to N,N'-dialkyl-and-diaryl terephthalaldimines and its stereochemistry contrary to behavior of other analogues

Article abstract of DOI:10.1080/10426507.2015.1091825

The addition of dimethyl phosphite to N,N'-terephthalylidene-alkyl- (or aryl-) amines resulted in new tetramethyl 1,4-phenylene-bis-(N-alkylaminomethyl)-phosphonates in moderate yields. The stereochemical behavior of these reactions was studied by NMR, de

Full text of DOI:10.1080/10426507.2015.1091825

Phosphorus, Sulfur, and Silicon and the Related Elements
ISSN: 1042-6507 (Print) 1563-5325 (Online) Journal homepage: http://www.tandfonline.com/loi/gpss20
Addition of Dimethyl Phosphite to N,N'-
dialkyl- and -diaryl Terephthalaldimines and Its
Stereochemistry Contrary to Behavior of Other
Analogues
Jarosław Lewkowski, Anna Ignaczak, Rafał Karpowicz, Justyna Obiedzińska &
Maria Rodriguez-Moya
To cite this article: Jarosław Lewkowski, Anna Ignaczak, Rafał Karpowicz, Justyna
Obiedzińska & Maria Rodriguez-Moya (2016): Addition of Dimethyl Phosphite to N,N'-
dialkyl- and -diaryl Terephthalaldimines and Its Stereochemistry Contrary to Behavior
of Other Analogues, Phosphorus, Sulfur, and Silicon and the Related Elements, DOI:
10.1080/10426507.2015.1091825
To link to this article: http://dx.doi.org/10.1080/10426507.2015.1091825
Accepted author version posted online: 19
Jan 2016.
View supplementary material
Submit your article to this journal
View related articles
Article views: 12
View Crossmark data
Full Terms & Conditions of access and use can be found at
http://www.tandfonline.com/action/journalInformation?journalCode=gpss20
Download by: [Orta Dogu Teknik Universitesi]
Date: 27 February 2016, At: 04:24

ACCEPTED MANUSCRIPT
Addition of Dimethyl Phosphite to N,N'-dialkyl- and -diaryl Terephthalaldimines
and Its Stereochemistry Contrary to Behavior of Other Analogues
Jarosław Lewkowski*a, Anna Ignaczak*b, Rafał Karpowicza, Justyna Obiedzińskaa, and Maria
Rodriguez-Moya
aDepartment of Organic Chemistry, Faculty of Chemistry, University of Łódź, Tamka 12,
91-403 Łódź, Poland
bDepartment of Theoretical and Structural Chemistry, Faculty of Chemistry, University of
Łódź, Pomorska 163/165, 91-236 Łódź, Poland
*E-mail: jlewkow@uni.lodz.pl; anignacz@uni.lodz.pl
Abstract: The addition of dimethyl phosphite to N,N’-terephthalylidene-alkyl- (or aryl-) amines
resulted in new tetramethyl 1,4-phenylene-bis-(N-alkylaminomethyl)-phosphonates in moderate
yields. The stereochemical behaviour of these reactions was studied by NMR demonstrating, that
the reactions lead to the formation of both possible diastereomeric forms in 1:1 up to 1:2 ratios.
This was unexpected considering that previously investigated additions of diethyl, dibenzyl and
diphenyl phosphites to terephthalaldimines were diastereoselective up to 100 %. Attempts were
made to find reasonable explanation for the phenomenon using semi-empirical and DFT
calculations.
1
ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
Keywords: Diaminophosphonates; terephthalic moiety; stereochemistry; dimethyl phosphite
2
ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
INTRODUCTION
During routine literature studies, we have found to our great surprise, that tetramethyl 1,4-
phenylene-di-(N-alkylaminomethylphosphonates) have been mentioned only once in the
literature and that this was nearly 50 years ago. Pudovik and Pudovik1 reported the synthesis of
two such compounds, but, as it was published in mid-sixties, no spectral characterisation had
been reported.
Therefore, in the name of scientific routine, we decided to synthesize these tetramethyl 1,4-
phenylene-di-(N-alkylaminomethylphosphonates) and to investigate pro forma the stereo-
chemistry of their formation. Our team investigated the addition of phosphorus nucleophiles to
azomethine bond in many aspects, inspired by the excellent work of Failla and Finocchiaro
group2-4 and by the studies of Gancarz and Tyka5. They found, that the addition of diethyl
phosphite to achiral terephthalic Schiff bases was diastereoselective up to 100 % and found also,
that this phenomenon was not easy to explain. It is to note, that these 1,4-phenylene-bis-(N-
alkylaminomethyl)-phosphonic acid derivatives have been found to have coordination abilities
towards Cu(II) ions6 or diaminophosphonate peptide receptor for lysine and arginine.7
We have discovered that the addition of diethyl, dibenzyl and diphenyl phosphites to achiral N-
alkyl terephthalic Schiff bases2-5,8 was diastereoselective to 100 %, but similar addition to N-aryl
ones was in some cases also highly stereoselective, while in other cases not stereoselective at all,
and it depended on the nature of the substituents at the aromatic ring9,10. The predominant or
exclusive diastereoisomer was found to adopt a meso-form.
Similar results have been reported for the addition of hypophosphorous acid to terephthalic
Schiff bases.11 This addition has been reported5,11 to be diastereoselective to 100 % for
3
ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
N-alkylaryl imines and to lead to a meso-form too, while the reactions performed on N-aryl
imines showed similar tendencies as described above.11 The surprising exception has been
noticed in this report,11 namely the addition of hypophosphorous acid to N-furfuryl terephthalic
Schiff base, did not show any evidence of diastereoselectivity, leading to the formation of both
diastereomeric forms in a 1:1 ratio.
Similar tendencies have been noticed in a case of the addition of di-(trimethylsilyl) phosphite to
achiral terephthalic Schiff bases, which led to the formation of 1,4-phenylene-bis-
(N-alkylaminomethyl)-phosphonic acids.12 The stereochemical behaviour of such processes
demonstrated, that - for several examples - this reaction led to the exclusive formation of only
one diastereomeric form, which was identified to be the racemic mixture.
When we thought that all is well explained, our routine examination of the tetramethyl 1,4-
phenylene-di-(N-alkylaminomethylphosphonate) formation revealed some astonishing facts,
which made our previous conclusions rather questionable. Addition of dimethyl phosphite to a
variety of achiral terephthalic Schiff bases was found to be either only slightly diastereoselective
or not diastereoselective at all!
RESULTS AND DISCUSSION
Terephthalaldimines 2a-h were prepared by the condensation of amines 1a-h with terephthalic
aldehyde in methanol at room temperature (Scheme 1). As this method is commonly
known,8,9,11,12 we found useless to report it here. Imines with N-alkylaryl (2a-c), N-aliphatic (2d)
and N-aryl groups (2e-h) were selected for the investigations, so that imines with all kinds of
N-substituents were studied.
4
ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
The preparation of tetramethyl 1,4-phenylene-bis(N-alkylaminomethyl)-phosphonates 3a-h was
carried out following the previously published procedures8,9,11,12 by the addition of dimethyl
phosphite to azomethine bond of terephthalic Schiff bases 2a-h in toluene for 5 hours (Scheme
1). Tetramethyl aminophosphonates 3a-h have been obtained as crystalline powder in moderate
to low yields. This is not unusual, as Barycki and coworkers5 suggested, that the conversion rate
for the addition to two azomethine groups is lower than that to one azomethine group.
Tetramethyl aminophosphonates 3a-h were purified by recrystallization and appropriate results
of spectroscopic and elemental analysis are given in the Experimental Part.
1
13
31
The H, C and in some cases P NMR spectroscopic data demonstrated, that all investigated
tetramethyl aminophosphonates 3a-h were formed with very low diastereoselectivities or even
with a complete lack of distereoselectivity. The formation of N-benzyl and N-furfuryl derivatives
was not stereoselective; NMR investigation showed the presence of the two diastereosisomers in
a 1:1 ratio. In the other cases investigated (formation of 3c and 3e-h) only a low stereoselectivity
ranging from 14 % de up to 43 % de was observed. Only the addition to N-t-butylaldimine 2d
was much more stereoselective giving diastereoisomeric products in 67 % de (Table 1). This
phenomenon was greatly unexpected in the light of previous results,2-5,8-9 which were discussed
in the introductory part. Additions of diethyl, dibenzyl and diphenyl phosphites led in most cases
to the exclusive formation of a meso-form.
The question appeared what might be the difference between these three phosphites and dimethyl
phosphite, so that the addition of the latter displays different stereochemistry. In order to put
light on the nature of this difference theoretical analysis was done.
5
ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
In previous papers8-12 we followed the Gancarz’s suggestion5 that addition of phosphites to
terephthalic Schiff bases proceeds through a two-step mechanism, where the first step is the
addition of a phosphite to the first azomethine bond resulting in the imino-aminophosphonate
derivative 4b. We have drawn a hypothesis8,9 that two molecules of imino-aminophosphonate
derivative 4b are able to form a dimer linked by hydrogen bonds. This dimer coordinates to
phosphite molecules and such a system is able to determine the stereochemistry of the addition to
the next azomethine bond to afford the meso form of bis-aminophosphonate 5 (Scheme 2).
However, considering the high similarity between diethyl phosphite and dimethyl phosphite, the
formation of bis-aminophosphonate 3a from the intermediate 4a should be characterised by a
similar stereoselectivity (Scheme 2). Taking into account, that the resolution of the problem in an
experimental way is extremely difficult (we did not manage to isolate any intermediate), we used
theoretical methods, which would help us to find the reason of such stereochemical behaviour. In
our computations, we followed Gancarz’s suggestion5 of a two-step reaction mechanism
mentioned above.
First we have performed PM6 and PM7 semi-empirical calculations for the addition of dimethyl
phosphite and diethyl phosphite to N,N’-terephthalilidenedibenzylamine. We computed the total
energies of equilibrium states of the structures formed at particular steps of both reactions, the
hypothetical iminoaminophosphonate intermediate 4a-b and 1,4-phenylene-bis-(N-
benzylaminomethylphosphonates) 3a and 5. Then the calculated values were compared. We have
chosen the addition to N,N’-terephthalilidenebenzylamine as a model for our computations
because of its typical behaviour and the relatively simple structure of the substrates.
6
ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
The calculations of the hypothetical dimethyl iminoaminophosphonate intermediate 4a were
performed separately for the cisoid and the transoid (with respect to the positions of the N
atoms) conformations, which, if differences in energies between them were high enough, might
play a key role in the stereochemistry of the reaction. The ground state structures found with the
PM6 method differ significantly from those obtained from the PM7 calculations, but the energy
differences between the most stable conformational states found with both methods for the cisoid
and the transoid form are so small, that they should be considered as insignificant. Similar
results were obtained for diethyl iminoaminophosphonate intermediate 4b. The differences
between the cisoid and the transoid forms of both 4a and 4b intermediates oscillate around 0.1
kcal / mol (Table 2). However, it is to note that in both cases, the transoid forms are more
privileged.
Semi-empirical methods are essentially approximate and may lead to erroneous representation of
reality. Therefore to verify these findings we applied a more advanced methodology, namely the
DFT method, using the B3LYP functional with a 6-31G(d,p) basis set. The energy differences
between cisoid and transoid conformations of the hypothetical imino amino-phosphonate
intermediates 4a and 4b at the B3LYP/6-31G(d,p) level of theory (see Table 2) appear to be even
smaller (0.05 kcal/mol) than those found with the semi-empirical methods. The DFT results
show also, that the structural properties of this system are somewhat better reproduced by the
PM6 method, but it predicts incorrectly the orientation of the phosphonate groups.
Since stability of the intermediates does not explain the phenomena observed in the experiment,
we performed similar study for the final products. The total energies of equilibrium states of
dimethyl and diethyl 1,4-phenylene-bis-(N-benzylaminomethyl-phosphonates) 3a and 5,
7
ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
respectively, were calculated for (R,S) and (S,S) diastereomers, as by definition the (R,R)
diastereomer behaves identically as the (S,S) one. As for the intermediates, the cisoid and
transoid forms were considered for each diastereoisomer. As can be seen in Table 3, all methods
predict that for both final products 3a and 5 there is no significant difference between the
energies of the most stable conformers of (S,S) and (R,S) isomers. One may note some
inconsistency in the lowest energy structures pointed out by the methods PM6 and PM7. The
other corresponding conformers (analogues of the B3LYP structures) appear to be higher in
energy by up to 1 kcal/mol.
Obviously, the above computations did not allow us to approach to the reason of the phenomena
observed experimentally. Therefore we turned our attention to substrates of the studied reaction.
We found only one difference, which we show in Table 4, namely both dimethyl and diethyl
phosphites show tendencies to form dimers and the dimer of dimethyl phosphite is more stable,
because its dissociation energy is higher by about 0.6 kcal/mol according to the DFT results. It is
then probable that the higher stability of the dimethyl phosphite dimer as compared to that of the
diethyl phosphite dimer plays a key role in explaining such a great difference in stereochemistry
of discussed reactions.
If we consider the previously suggested mechanism8-12 to be true, such a difference between
tendencies of dimer formation of both phosphites should be critical for the formation of active
complexes AC-4a and AC-4b (Figure 1) Because the diethyl phosphite dimer is less stable, the
formation of the active complex AC-4b is much more probable, while the higher stability of
dimethyl phosphite dimer may disfavour the formation of the AC-4a complex. However, it is to
stress that the formation of active complexes AC-4a and AC-4b is hypothetical. We did not
8
ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
succeed in isolating intermediates 4a,b, the characterization of which would confirm
experimentally the formation of complexes AC-4a and AC-4b. Also, considering the
discrepancies between the semiempirical and DFT results presented in Tables 2-4, as well as the
large number of variables involved, verification of our hypothesis at a reliable theoretical level
would be a very long task, therefore could not be done at the moment.
Summarizing, the thorough analysis performed in the present work allowed us to eliminate the
possibility that different stability of single intermediate or final products is responsible for the
phenomenon of different diastereoselectivity. We managed however to find another factor that
hypothetically could be its source. An important achievement of our study is also the detailed
knowledge of specific properties of the compounds participating in these reactions, which
constitutes a good starting point for further investigations.
EXPERIMENTAL
All solvents (POCh-Poland) were routinely distilled and dried prior to use. Amines, dimethyl
phosphite and terephthalic aldehyde (Aldrich) were used as received. NMR spectra were
recorded with a Bruker Avance III 600 MHz apparatus operating at 600 MHz (1H), 150 MHz
(13C) and 243 MHz (31P). Elemental analyses were performed in the Centre for Molecular and
Macromolecular Science of the Polish Academy of Science in Łódź. Melting points were
measured on a MelTemp II apparatus and are not corrected. The Supplemental Materials
contains sample 1H, 13C and 31P NMR spectra for 3a-3d (Figures S 1 – S 11)
Tetramethyl 1,4-phenylene-bis-(aminomethylphosphonates) (3a-h). General procedure.
Terephthalic aldehyde (1.34 g, 10 mmol) was dissolved in methanol (40 mL) and then the
corresponding amine (20 mmol) was added. The mixture was stirred at room temperature for 24
9
ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
h and the precipitated solid was then collected and dried. The thus obtained Schiff base 2a-f was
used for the next step without further purification.
The Schiff base (2a-f) (5 mmol) was dissolved in toluene (30 mL) and then the dimethyl
phosphite (10 mmol, 1.10 g) was added. The mixture was refluxed for 5 to 7 h. After cooling to
room temperature the solid precipitate was collected by filtration, washed, and dried. Products
were purified by recrystallization. A separation of the diastereoisomers via column
chromatography or crystallization failed.
Tetramethyl 1,4-phenylene-bis-(N-benzylaminomethylphosphonate) (3a)
Yield: 85 % (2.13 g); mp: 103-104 C (dichloromethane). 1H NMR (CDCl3, 600 MHz):  7.26
2
(s, 4H, C6H4); 7.14-7.11 (m, 4 H, PhH); 7.08-7.07 (m, 6 H, PhH); 3.90 (d, JPH = 19.4 Hz, 1H,
CHP) and 3.89 (d, 2JPH = 19.6 Hz, 2H, CHP), (1:1); 3.63 (d, 2JHH = 13.3 Hz, 2H, CHAHB); 3.56
3
3
2
(d, JPH = 10.5 Hz, 3H, P-O-CH3), 3.55 (d, JPH = 10.5 Hz, 3H, P-O-CH3); 3.38 (d, JHH = 13.3
Hz, 2H, CHAHB); 3.36 (d, 3JPH = 10.5 Hz, 3H, P-O-CH3), 3.35 (d, 3JPH = 10.5 Hz, P-O-CH3, 3H).
13C NMR (CDCl3, 150 MHz):  139.2 (Cipso); 135.5 (Cipso); 128.9 (Cpara); 128.4 (Carom); 128.2
1
(Carom); 127.2 (Carom); 59.1 (d, JPC = 153.2 Hz, CP); 53.6 (m, COP); 53.3 (m, COP); 51.32 (d,
3
3JPC = 9.0 Hz, CNCP); 51.26 (d, JPC = 9.0 Hz, CNCP). 31P NMR (CDCl3, 243 MHz):  = 25.6.
Anal. Calcd. for C26H34N2O6P2: C, 58,64; H, 6,44; N, 5,26. Found: C, 58.46; H, 6.51; N, 5.23 %.
Tetramethyl 1,4-phenylene-bis-(N-furfurylaminomethylphosphonate) (3b)
Yield: 61 % (1.49 g); mp: 103-107 C (dichloromethane). 1H NMR (CDCl3, 600 MHz):  7.44
(s, 4H, C6H4); 7.35-7.34 (m, 2H, CH5fur); 6.29 (dd, J = 1.5 Hz, 3.0 Hz, 2H, H4fur); 6.11-6.10 (m,
2H, H3fur); 4.10 (d, 2JPH = 18.4 Hz, 2H, CHP), 4.09 (d, 2JPH = 18.4 Hz, 2H, CHP); 3.81 (d, 2JHH
=
3
2
14.3 Hz, 2H, CHAHB); 3.71 (d, JPH = 10.6 Hz, 6H, P-O-CH3); 3.59 (d, JHH = 14.3 Hz, 2H,
10
ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
3
3
13
CHAHB); 3.58 (d, JPH = 10.2 Hz, 3H, P-O-CH3), 3.57 (d, JPH = 10.2 Hz, 3H, P-O-CH3). C
NMR (CDCl3, 150 MHz):  = 152.8 (Cipso); 142.0 (Cfur); 135.1 (Cipso); 128.9 (Cpara); 110.1 (Cfur);
107.7 (Cfur); 58.86 (d, 1JPC = 153.5 Hz, CP); 58.83 (d, 1JPC = 153.5 Hz, CP); 53.53 (d, 2JPC = 35.5
Hz, COP); 53.51 (d, 2JPC = 35.5 Hz, COP); 43.8 (d, 3JPC = 8.8 Hz, CNCP); 43.7 (d, 3JPC = 8.9 Hz,
CNCP). 31P NMR (CDCl3, 243 MHz):  = 25.3 (br s). Anal. Calcd. for C22H30N2O8P2: C, 51,57;
H, 5,90; N, 5,47. Found: C, 51.59; H, 6.10; N, 5.31 %.
Tetramethyl 1,4-Phenylene-bis-(N-benzhydrylaminomethylphosphonate) (3c)
Yield: 53 % (1.91 g); mp: 181-183 C (toluene / hexane 1:1). 1H NMR (CDCl3, 600 MHz): 
= 7.36-7.33 (m, 12H, Ph, C6H4); 7.29-7.24 (m, 10H, Ph); 7.21-7.19 (m, 2H, Ph); 4.71 (s, 1H,
2
2
CHPh2); 4.69 (s, 1H, CHPh2); 4.02 (d, JPH = 21.8 Hz, 1H, CHP), 4.00 (d, JPH = 21.8 Hz, 2H,
3
3
CHP); 3.87 (d, JPH = 10.6 Hz, 3H, P-O-CH3); 3.86 (d, JPH = 10.6 Hz, 3H, P-O-CH3); 3.52 (d,
3
3JPH = 10.1, 3H, P-O-CH3); 3.51 (d, JPH = 10.6 Hz, 3H, P-O-CH3). 13C NMR (CDCl3, 150
MHz):  = 143.5 (Cipso); 141.8 (Cipso); 135.6 (Ctereph); 128.9 (m, Cipso); 128.7 (Carom); 128.5
3
(Carom); 127.8 (Carom); 127.5 (Carom); 127.24 (Carom); 127.21 (Carom); 63.9 (d, JPC = 8.7 Hz,
CNCP); 63.9 (d, 3JPC = 8.5 Hz, CNCP); 57.5 (d, 1JPC = 154.5 Hz, CP); 53.9 (m, COP); 53.2 (m,
31
COP). P NMR (CDCl3, 243 MHz):  = 25.9 (br s). Anal. Calcd. for C38H42N2O6P2: C, 66,66;
H, 6,18; N, 4,09. Found: C, 66.33; H, 6.26; N, 3.83 %.
Tetramethyl 1,4-Phenylene-bis-(N-t-butylaminomethylphosphonate) (3d)
1
Yield: 57 % (1.25 g); mp: 145-149 C (dichloromethane). H NMR (CDCl3, 600 MHz): 
7.15 (s, 4H, C6H4); 4.18 (d, 2JPH = 24.1 Hz, 1H, PCH); 4.17 (d, 2JPH = 24.1 Hz, 2H, PCH); 3.79
3
3
3
(d, JPH = 10.2 Hz, 3H, P-O-CH3); 3.78 (d, JPH = 10.2 Hz, 3H, P-O-CH3); 3.41 (d, JPH = 10.1
Hz, 3H, P-O-CH3); 3.40 (d, 3JPH = 10.1 Hz, 3H, P-O-CH3); 1.01 (s, 9H, C(CH3)3); 0.98 (s, 18H,
11
ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
13
C(CH3)3). C NMR (CDCl3, 150 MHz):  = 139.2 (Cipso); 139.0 (Cipso); 128.3 (m, Ctereph); 55.3
(d, 1JPC = 155.2 Hz, CP); 55.2 (d, 1JPC = 155.5 Hz, CP); 54.4 (m, COP); 53.2 (m, COP); 52.4 (d,
3JPC = 8.2 Hz, CNCP); 52.3 (d, 3JPC = 7.8 Hz, CNCP); 29.9 (CH3); 29.8 (CH3). 31P NMR (CDCl3,
243 MHz):  = 26.3, 25.5 (5:1). Anal. Calcd. for C20H38N2O6P2: C, 51,72; H, 8,25; N, 6,03.
Found: C, 51.98; H, 8.38; N, 5.83 %.
Tetramethyl 1,4-Phenylene-bis-(N-(p-methylphenyl)aminomethylphosphonate) (3e)
1
Yield: 35 % (0.84 g); mp: 197-200 C (dichloromethane). H NMR (CDCl3, 600 MHz):
7.44 (s, 4H, C6H4); 6.89 (A part of AA'XX' spectrum, 3J = 8.3 Hz, 4J = 3.4 Hz, 4H); 6.49 (X
2
part of AA'XX' spectrum, 4H); 4.75 and 4.74 (2d, JPH = 22.9 and 23.4 Hz, PCH, 2H); 3.73 and
3.72 (2d, 3JPH = 10.6 Hz, P-O-CH3, 6H); 3.42 and 3.34 (2d, 3JPH = 10.5 Hz, P-O-CH3, 6H); 2.18
and 2.17 (2s, CH3, 6H). 13C NMR (CDCl3, 150 MHz):  143.65 (m, Cipso); 135.74 (Ctereph);
3
129.60 (Carom); 129.57 (Carom); 128.20 (d, JCP = 7.5 Hz, CipsoNCP); 127.87 (Carom); 127.86
(Carom); 114.10 (Carom); 114.06 (Carom); 55.83 (d, 1JCP = 150.7 Hz, CP); 55.78 (d, 1JCP = 150.4 Hz,
CP); 53.75 (m, COP); 53.63 (m, COP); 20.27 (CH3); 20.26 (CH3). 31P NMR (CDCl3, 243 MHz):
1
 24.90 and 24.86 (3:4). Anal. Calcd. for C26H34N2O6P2  /4CH2Cl2: C, 56,94; H, 6,28; N, 5,06.
Found: C, 56.82; H, 6.00; N, 4.90 %.
Tetramethyl 1,4-Phenylene-bis-(N-(p-methoxyphenyl)aminomethylphosphonate) (3f)
1
Yield: 16 % (0.4 g); mp: 195-197 C (dichloromethane). H NMR (CDCl3, 600 MHz):  =
7.47 (s, 4H, C6H4); 7.45 (s, 4H, C6H4); 6.68-6.65 (A-part of AA'XX' spectrum, 2H); 6.58-6.55
2
2
(X-part of AA'XX' spectrum, 2H); 4.72 (d, JPH = 24.0 Hz, 1H, PCH); 4.71 (d, JPH = 24.0 Hz,
3
1H, PCH); 3.74 (d, 3JPH = 10.8 Hz, 3H, P-O-CH3); 3.71 (d, JPH = 10.8 Hz, 3H, P-O-CH3); 3.69
(s, OCH3, 3H); 3.68 (s, OCH3, 3H); 3.41 (d, 3JPH = 10.8 Hz, 3H, P-O-CH3); 3.36 (d, 3JPH = 10.8
12
ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
13
3
Hz, 3H, P-O-CH3) (4:9). C NMR (DMSO-D6, 150 MHz):  = 151.4 (Cipso); 140.8 (d, JPC
=
15.0 Hz, CipsoNCP); 135.9 (Carom); 127.9 (Carom); 114.63 (Carom); 114.59 (Carom); 114.2 (Carom);
1
1
114.10 (Carom); 114.06 (Carom); 55.1 (OCH3); 54.04 (d, JPC = 151.5 Hz, CP); 53.97 (d, JPC
=
31
151.4 Hz, CP); 53.2 (m, COP); 52.8 (m, COP). P NMR (CDCl3, 243 MHz):  = 24.6. Anal.
Calcd. for C26H34N2O8P2: C, 55,32; H, 6,07; N, 4,96. Found: C, 55.11; H, 6.11; N, 4.81 %.
Tetramethyl 1,4-Phenylene-bis-(N-(m-methylphenyl)aminomethylphosphonate) (3g)
Yield: 35 % (0.96 g); mp: 171-174 C (dichloromethane / hexane 1:1). 1H NMR (CDCl3, 600
MHz):  = 7.46 (s, 4H, C6H4); 6.97-6.94 (s, 2H, m-C6H4); 6.52 (dd, J = 7.8 Hz, 7.2 Hz, 2H, m-
C6H4); 6.40 (br s, 2H, m-C6H4); 6.36 (dd, 3J = 7.8 Hz, 4J = 2.6 Hz, 2H, m-C6H4); 4.78 (d, 2JPH
23.3 Hz, 1H, PCH); 4.77 (d, 2JPH = 23.3 Hz, 1H, PCH); 3.73 (d, 3JPH = 10.6 Hz, 3H, P-O-CH3);
=
3
3
3
3.72 (d, JPH = 10.6 Hz, 3H, P-O-CH3); 3.42 (d, JPH = 10.5 Hz, 3H, P-O-CH3); 3.35 (d, JPH
=
10.5 Hz, 3H, P-O-CH3); 2.19 (s, 3H, CH3); 2.18 (s, 3H, CH3). 13C NMR (CDCl3, 150 MHz):  =
146.0 (m, Cipso); 138.91 (Cipso); 138.88 (Cipso); 135.7 (Ctereph); 128.97 (Carom); 128.94 (Carom);
128.2 (m, CipsoNCP); 119.53 (Carom); 119.50 (Carom); 114.77 (Carom); 114.74 (Carom); 110.94
(Carom); 110.91 (Carom); 55.50 (d, 1JPC = 150.6 Hz, CP); 55.45 (d, 1JPC = 150.0 Hz, CP); 53.7 (m,
COP); 21.4 (CH3). 31P NMR (CDCl3, 243 MHz):  = 24.83; 24.78 (5:2). Anal. Calcd. for
C26H34N2O6P2: C, 58,64; H, 6,44; N, 5,26. Found: C, 58.37; H, 6.51; N, 5.34 %.
Tetramethyl 1,4-Phenylene-bis-(N-(m-methoxyphenyl)aminomethylphosphonate) (3h)
1
Yield: 62 % (1.75 g); mp: 145-148 C (toluene). H NMR (CDCl3, 600 MHz):  = 7.45 (s,
4H, C6H4); 6.99-6.96 (m, 2H, m-C6H4); 6.25-6.23 (m, 2H, m-C6H4); 6.19-6.17 (m, 2H, m-C6H4);
2
2
6.09-6.08 (m, 2H, m-C6H4); 4.77 (d, JPH = 23.4 Hz, 1H, PCH); 4.75 (d, JPH = 23.4 Hz, 1H,
3
3
PCH); 3.73 (d, JPH = 10.6 Hz, 3H, P-O-CH3); 3.72 (d, JPH = 10.6 Hz, 3H, P-O-CH3); 3.66 (s,
13
ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
3H, OCH3); 3.65 (s, 3H, OCH3); 3.40 (d, 3JPH = 10.4 Hz, 3H, P-O-CH3); 3.35 (d, 3JPH = 10.4 Hz,
13
3H, P-O-CH3). C NMR (CDCl3, 150 MHz):  = 160.7 (Cipso); 147.3 (m, Cipso); 135.7 (Ctereph);
129.90 (Carom); 129.88 (Carom); 128.2 (m, CipsoNCP); 106.93 (Carom); 106.91 (Carom); 103.84
(Carom); 103.79 (Carom); 100.17 (Carom); 100.13 (Carom); 55.57 (d, 1JPC = 150.2 Hz, CP); 55.53 (d,
31
1JPC = 150.2 Hz, CP); 55.0 (OCH3); 53.8 (m, COP). P NMR (CDCl3, 243 MHz):  = 24.62;
24.59 (2:3). Anal. Calcd. for C26H34N2O8P2: C, 55,32; H, 6,07; N, 4,96. Found: C, 55.29; H,
6.27; N, 4.76 %.
Quantumchemical Calculations
The first series of our calculations has been performed using the MOPAC2012 program.13
The structures of dimethyl and diethyl iminoaminophosphonate intermediates 4a,b, as well as
tetramethyl and tetraethyl 1,4-phenylene-bis-(N-benzylamino-methylphosphonates) 3a and 5,
were fully optimized (no constraints) at the semi-empirical levels PM6 and PM7. No symmetry
was imposed on the molecules geometry (C1) in all calculations. The potential energy surface of
these systems is expected to have many local minima, but the systematic scan of all possible
conformers is impossible because of an excessive number of torsional angles in these structures.
Therefore the following strategy was used to find the most stable structures: first a large number
of various conformers (taking into account rotations of different parts of the molecules around
bonds C-C, C-P or C-N) were tested as starting structures for the PM6 and PM7 optimization.
The lowest energy structures were then used as starting geometries in the conformational search
performed using the stochastic dynamic method via the Verlet velocity algorithm, available in
the program Gabedit.14 In the simulation the conformational space was explored at a certain
simulation time and temperature. During this search 50 lowest energy structures were selected
14
ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
and saved, and subsequently optimized. The procedure was repeated with different simulations
conditions: for two different friction parameters of 40 and 80 ps-1, as well as for different
temperatures, which initially were set to 1000 K, but for the lowest energy structures found in the
simulations an additional verification was made by performing a subsequent simulation in 500 K
and in 300 K. The simulation time for small molecules and at the verification steps was usually
set to 5-10 ps, while for larger structures, such as 3a and 5, it was increased up to 50-70 ps. In all
simulations the timesetp of 1 fs was used. The procedure was repeated independently with the
PM6 and PM7 methods.
At the second stage of this part of our work the full geometry optimization of the same
compounds was repeated at the more accurate level using the density functional method
B3LYP15-17 combined with the basis set 6-31G(d,p). However, due to a relatively long
computational time, these calculations were limited to the lowest energy structures indicated by
the semi-empirical PM6 and PM7 results for two isomers (cisoid and transoid conformers with
respect to the positions of atoms N) of the intermediates 4a-b and for four isomers (cisoid (R,S),
cisoid (S,S), transoid (R,S), and transoid (S,S)) of the final products 3a and 5. Additionally, for
the final products, which were the most problematic, for each lowest energy structure found from
the B3LYP optimization we tested all corresponding conformers, created by applying the
appropriate symmetry operations. Some additional tests were performed taking into account an
effect of rotation of methyl and ethyl groups. The same strategy was used for the dimers of
dimethyl and diethyl phosphites. All DFT calculations were performed with the Gaussian 09
package.18
15
ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
At both stages for the resulting lowest energy structures the vibrational analysis was performed
to verify whether these geometries correspond to true energy minima, which was confirmed by
the absence of imaginary frequencies. The pictures of the optimized structures shown in Tables
2-4 were made with the program Gabedit.14
ACKNOWLEDGEMENTS
Small financial support of the Faculty of Chemistry, University of Łódź is kindly acknowledged.
Quantum calculations were supported in part by PL-Grid Infrastructure.
REFERENCES
1. Pudovik, A. N.; Pudovik, D. A. J. Gen. Chem. USSR 1966, 36, 1467-1472.
2. Failla, S.; Finocchiaro, P. Phosphorus, Sulfur Silicon Relat. Elem., 1993, 85, 65-72.
3. Failla, S.; Finocchiaro, P.; Hägele, G.; Rapisardi, R. Phosphorus, Sulfur Silicon Relat.
Elem., 1993, 82, 79-90.
4. Failla, S.; Finocchiaro, P. Phosphorus, Sulfur Silicon Relat. Elem., 1995, 107, 79-86.
5. Barycki, J.; Gancarz, R.; Milewska, M.; Tyka, R. Phosphorus, Sulfur Silicon Relat. Elem.,
1995, 105, 117-122.
6. Gałęzowska J.; Szyrwiel Ł.; Młynarz P.; Śliwińska S.; Kafarski P.; Kozłowski H.
Polyhedron 2007, 26, 4287-4293.
7. Młynarz P.; Olbert-Majkut A.; Śliwińska S.; Schroeder G.; Bańkowski B.; Kafarski P. J.
Mol. Struct. 2008, 873, 173-180.
8. Lewkowski, J.; Rzeźniczak, M.; Skowroński, R. Heteroatom Chem. 2000, 11, 144-151.
9. Lewkowski, J. Phosphorus, Sulfur Silicon Relat. Elem., 2005, 180, 179-195.
16
ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
10. Lewkowski, J.; Stronka-Lewkowska, E. Phosphorus, Sulfur Silicon Relat. Elem., 2006, 181,
1323-1330.
11. Lewkowski J., Rybarczyk, M. Heteroatom Chem. 2008, 19, 283-287.
12. Lewkowski, J.; Dzięgielewski, M. Heteroatom Chem. 2009, 20, 431-435.
13. MOPAC2012, Stewart, J. J. P. Stewart Computational Chemistry, Colorado Springs, CO,
USA, HTTP://OpenMOPAC.net, 2012.
14. Gabedit, Allouche, A. R. J. Comput. Chem. 2011, 32, 174-182
15. Becke, A. D. J. Chem. Phys. 1993, 98, 5648-5652.
16. Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37, 785-789.
17. Miehlich, B.; Savin, A.; Stoll, H.; Preuss, H. Chem. Phys. Lett. 1989, 157, 200-206.
18. Gaussian 09, Revision A.02, Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.;
Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.;
Sonnenberg, J. L.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.;
Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A.,Jr.; Peralta,
J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin, K. N.; Staroverov, V. N.;
Kobayashi, R.; Normand, J.; Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar, S. S.;
Tomasi, J.; Cossi, M.; Rega, N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken,
V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.;
Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.; Morokuma, K.; Zakrzewski, V. G.;
Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, O.;
17
ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian, Inc., Wallingford CT,
2009.
18
ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
Table 1 Results of the addition of dimethyl phosphite to azomethine bond of imines 2a-h
Yield Diastereomeric Diastereomeric
R
3
a
b
c
d
e
f
[%]
ratio (dr)
excess (de) [%]
CH2Ph
85
1:1
--
CH2Fur
61
53
57
35
16
35
62
1:1
2:1
5:1
3:4
4:9
5:2
2:3
--
CHPh2
33
67
14
39
43
20
C(CH3)3
p-CH3-C6H4
p-OCH3-C6H4
m-CH3-C6H4
m-OCH3-C6H4
g
h
19
ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
Table 2 PM6, PM7 and DFT B3LYP/6-31G(d,p) total energies and ZPE corrected energies of
equilibrium states of hypothetical intermediates 4a and 4b for cisoid and transoid forms
Form
cisoid dimethyl 4a transoid dimethyl 4a
cisoid diethyl 4b
transoid diethyl 4b
E
-107998.56
-107725.81
-107998.69
-107725.92
-114917.90
-114612.18
-114918.01
-114612.26
[kcal/mol]
E+ZPE
[kcal/mol]
E
-109465.99
-109188.48
-109466.10
-109189.02
-116390.24
-116079.29
-116390.34
-116079.88
[kcal/mol]
E+ZPE
[kcal/mol]
E
-1008670.552
-1008376.24
-1008670.56
-1008376.19
-1058021.10
-1057691.15
-1058021.15
-1057691.12
[kcal/mol]
E+ZPE
[kcal/mol]
20
ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
Table 3 PM7 and DFT B3LYP/6-31G(d,p) total energies of equilibrium states of tetramethyl
and tetraethyl 1,4-phenylene-bis-(N-benzylamino-methylphosphonates) 3a and 5 for (R,S) and
(S,S) diastereomers, respectively
Isomer
(S,S) tetramethyl
(R,S) tetramethyl
(S,S) tetraethyl 5 (R,S) tetraethyl 5
3a
3a
E
-139382.01
-139049.67
-139382.16
-139049.71
-153220.99
-152822.66
-153221.01
-152822.59
[kcal/mol]
E+ZPE
[kcal/mol]
E
-141842.57
-141505.72
-141842.37
-141505.91
-155690.74
-155287.61
-155690.55
-155287.40
[kcal/mol]
E+ZPE
[kcal/mol]
E
-1414996.23
-1414635.38
-1414996.16
-1414635.39
-1513697.2808
-1513265.1797
-1513697.3117
-1513265.0813
[kcal/mol]
E+ZPE
[kcal/mol]
21
ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
Table 4 DFT B3LYP/6-31G(d,p) total energies of dimethyl and diethyl phosphites of a single
and a dimeric form. Edys was calculated according to: Edys (ZPE)= 2xEsingle - Edimer from values
including the ZPE corrections.
Diethyl
phosphite
(dimer)
Form of
a substrate
Dimethyl
phosphite (single)
Dimethyl phosphite Diethyl phosphite
(dimer)
(single)
E [kcal/mol]
E+ZPE
[kcal/mol]
Edys (E+ZPE)
[kcal/mol]
-31343.10
-31286.29
-62709.55
-62594.10
-38262.45
-38172.62
-76547.56
-76366.27
21.52
21.03
E [kcal/mol]
E+ZPE
[kcal/mol]
Edys (E+ZPE)
[kcal/mol]
-32368.58
-32311.96
-64743.35
-64629.64
-39292.91
-39202.87
-78590.79
-78410.50
5.72
4.76
E [kcal/mol]
E+ZPE
[kcal/mol]
Edys (E+ZPE)
[kcal/mol]
-406310.24
-406247.02
-812628.77
-812501.43
-455661.23
-455562.25
-911329.96
-911131.27
7.39
6.77
22
ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
Figure 1 Structures of the active complexes AC-4a and AC-4b
23
ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
Scheme 1 Reaction of terephthalaldimines 2a - h with dimethyl phosphite
24
ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
Scheme 2 Formation of the bis(aminophosphonates) 3a and 5
25
ACCEPTED MANUSCRIPT