A simple and efficient catalyst system for the asymmetric transfer hydrogenation of ketones

Article abstract of DOI:10.1055/s-2007-986653

Aryl alkyl ketones are efficiently and selectively reduced (up to 97% ee) under transfer-hydrogenation conditions in 2-propanol using rhodium catalysts based on readily available amino acid derived hydroxamic acid ligands. Georg Thieme Verlag Stuttgart.

Full text of DOI:10.1055/s-2007-986653

LETTER
2541
A Simple and Efficient Catalyst System for the Asymmetric Transfer
Hydrogenation of Ketones
R
hodium-Catalyze
a
d
A
symm
e
t
tric Tra
r
nsfer Hy
i
drogen
n
ation Ahlford, Alexey B. Zaitsev, Jesper Ekström, Hans Adolfsson*
Department of Organic Chemistry, Stockholm University, the Arrhenius Laboratory, 106 91 Stockholm, Sweden
Fax +46(8)154908; E-mail: hansa@organ.su.se
Received 11 July 2007
The major topological difference between the two classes
of ligands presented in Figure 1 is their potential dentic-
ities. Whereas the pseudo-dipeptides have the possibility
to coordinate the metal with the carbamate, the central
amide and the hydroxyl functionality, the thioamides can
only act as bidentate ligands. We know from previous
studies that in the case of the pseudo-dipeptides, all three
functionalities are of utmost importance for the activity of
the catalytic system, and furthermore, the possibility to
deprotonate the central amide of the ligand is most crucial.
To increase stability of the catalysts derived from the
pseudo-dipeptides we focused on improving their coordi-
nating ability. An obvious change was to decrease the pKa
of the central amide, thereby facilitating an easy deproto-
nation which would result in stronger coordination. Thio-
amides are considerably more acidic than the
corresponding amides and exchanging oxygen for sulfur
in the central amide functionality of a pseudo-dipeptide
resulted in a ligand with significantly different proper-
ties.4 When a ruthenium complex containing this thio-
pseudo-dipeptide ligand was employed as catalyst in
ATH, a switch of product enantioselectivity was ob-
served. Hence, employing catalysts based on pseudo-
dipeptide ligands derived from L-amino acids typically re-
sulted in the formation of enantioenriched S-alcohols,
whereas when the corresponding thioamide ligand was
used, the R-alcohol was the major isomer. Moreover, we
found that the alcohol functionality was no longer re-
quired and good catalytic activity and selectivity was ob-
tained with the structurally simpler compound.
Abstract: Aryl alkyl ketones are efficiently and selectively reduced
(up to 97% ee) under transfer-hydrogenation conditions in 2-pro-
panol using rhodium catalysts based on readily available amino acid
derived hydroxamic acid ligands.
Key words: asymmetric catalysis, hydrogen transfer, reductions,
rhodium, ketones
Asymmetric transfer hydrogenation (ATH) of ketones is a
convenient method for the preparation of chiral secondary
alcohols.1 The method employs inexpensive and safe
hydrogen donors such as simple secondary alcohols (e.g.,
2-propanol) or formic acid (most often as sodium formate
or the formic acid triethylamine azeotrope), instead of
more hazardous and less conveniently handled molecular
hydrogen. A significant progress in the field of ATH with
respect to the development of new catalysts and ligands
was achieved during the last decade. The introduction of
catalysts based on ruthenium, rhodium, or iridium result-
ed in major improvements concerning the activity and
selectivity of the reduction process.2
We have previously reported on pseudo-dipeptides and
thioamides derived from Boc-protected amino acids as ef-
ficient ligands for the ATH of aromatic ketones employ-
ing Ru or Rh half-sandwich complexes (Figure 1).3–5
These ligands feature no intrinsically basic nitrogen cen-
ters which is the case with the classical ligands used in this
process, i.e. amino alcohols and monosulfonated di-
amines.
To investigate whether this enantioswitchable behavior is
a consequence of the ligand acidity we decided to affect
the acidity of the amide NH bond by changing the substi-
tution pattern on the amide nitrogen rather than replacing
the carbonyl oxygen by sulfur. We figured that the intro-
duction of an electron-withdrawing substituent on the
amide nitrogen would lead to a more acidic NH bond, a
situation similar to the thioamide case, and the choice fell
on one of the most simple derivatives fulfilling this crite-
ria, namely the hydroxamic acid. In the field of asymmet-
ric catalysis, hydroxamic acids have been successfully
employed predominantely in oxidation reactions using
early transition metals (e.g. vanadium).6 However, herein
we report on the efficient Rh-catalyzed ATH of ketones
employing a novel set of N-Boc-protected amino acid
ligands containing the hydroxamic acid functionality.
O
R2
∗
R1
R3
∗
S-selective
N
H
NH
R1
OH
Boc
S
R2
N
H
R-selective
NH
Boc
Figure 1 Stereochemical outcome of ATH catalyzed by Ru or Rh
complexes of pseudo-dipeptide and thioamide ligands, respectively
SYNLETT 2007, No. 16, pp 2541–2544
Advanced online publication: 12.09.2007
0
1
.1
0
.2
0
0
7
DOI: 10.1055/s-2007-986653; Art ID: G24207ST
© Georg Thieme Verlag Stuttgart · New York

2542
K. Ahlford et al.
LETTER
Hydroxamic acid ligands 1a–d were prepared using the
one-pot procedure introduced by Giacomelli and co-
workers, starting from the appropriate N-Boc-protected
amino acids (Scheme 1).7 Unfortunately, we were not able
to reproduce the yields reported in the Giacomelli paper,
and in our hands the method gave a maximum yield of
40%, even though we were using a modified procedure
provided to us by the authors.8
Table 1 N-Hydroxyl Amides of N-Boc-Protected Amino Acids as
Ligands for the Rhodium-Catalyzed ATH of Acetophenone in
2-Propanola
O
OH
[RhCl2Cp*]2
L*, i-PrONa
i-PrOH
Entry
Ligand LiCl
Time
Conv.
(%)b
ee
(5 mol%) (h)
(%, conf.)b
Cl
N
Cl
1
2
1a
1a
1a
1b
1b
1b
1b
1b
1c
1c
1d
–
+
–
–
–
+
+
–
–
+
–
0.5
0.5
0.5
0.5
2
66
56
4
82 (S)
88 (S)
71 (S)
85 (S)
87 (S)
97 (S)
97 (S)
87 (S)
86 (S)
92 (S)
12 (S)
, NMM
N
N
O
O
R
OH
R
Cl
N
H
OH
3c
4
DMAP, NH2OH·HCl
NH
NH
Boc
Boc
CH2Cl2, r.t.
55
89
45
82
89
45
63
95
O
O
5
OH
OH
OH
N
N
H
6
0.5
2
H
NH
NH
NH
Boc
Boc
1a
1b
7
O
O
8d
9
2
OH
Ph
Ph
Boc
N
H
N
H
2
NH
Boc
1c
1d
10
11
2
Scheme 1 One-pot procedure for the formation of hydroxamic acid
2
ligands
a Reaction conditions: acetophenone (1 equiv, 0.2 M in i-PrOH),
[{RhCl2Cp*}2] (0.25 mol%), ligand (0.55 mol%) and i-PrONa (5
mol% unless otherwise indicated), r.t.
Nevertheless, the hydroxamic acid ligands 1a–d were
evaluated in the rhodium-catalyzed ATH of acetophenone
in 2-propanol using 0.5 mol% catalytic loading (Table 1).
The rhodium catalysts were formed in situ by mixing
[RhCl2Cp*]2, ligand and sodium isopropoxide in 2-pro-
panol. The amount of base turned out to be crucial and us-
ing a ratio of base to catalyst below 3 gave no conversion
to the product.9 Increasing the amount of base to 5 mol%
resulted, however, in high conversions and ee using hy-
droxamic acids 1a–c (Table 1, entries 1, 5, and 9).10 Most
interestingly, in contrast to ATH reactions performed with
catalysts based on thioamide ligands (Figure 1), we
found that 1-phenylethanol was formed with opposite
configuration employing these simple hydroxamic acid
ligands. The turnover frequency was slightly higher for
the catalyst derived from ligand 1a (TOF = 264 h–1) as
compared to catalysts formed with ligands 1b and 1c
(TOF = 220 and 180 h–1, respectively).11 The selectivity
was on the other hand somewhat better using ligands 1b
or 1c. The use of the phenylglycine-derived ligand 1d
gave good conversion, although with low enantioselectiv-
ity (Table 1, entry 11). We have previously reported that
addition of lithium chloride as a co-catalyst to the ATH
system can have beneficial effects on both catalyst activi-
ty and selectivity.3d,4 Accordingly, the addition of LiCl (5
mol%) to the reaction mixture led to a substantial increase
of the reaction performance, and up to 97% ee was ob-
tained for 1-phenylethanol using the valine-derived ligand
1b (Table 1, entry 6). Interestingly, when the [RuCl2(p-
cymene)]2 complex, normally an efficient precatalyst
b Conversion and enantioselectivity were determined by GLC (CP
Chirasil DEXCB).
c [{RuCl2(p-cymene)}2] (0.25 mol%) was employed as catalyst
precursor.
d The substrate was added 10 min after mixing the other reaction
components.
together with amino alcohol, monotosylated diamine or
pseudo-dipeptide ligands in the ATH of aryl alkyl ke-
tones, was employed as a metal source, a considerably
poorer result was obtained (Table 1, entry 3).
From the results obtained in the ligand screening we
found that the catalytic system containing the Rh catalyst
based on the valine-derived ligand 1b in combination with
LiCl as co-catalyst showed overall superior activity and
selectivity. Employing this catalyst mixture we studied
the scope of the reaction and the results are presented in
Table 2. As seen in Table 2, a range of differently substi-
tuted acetophenones were efficiently reduced with up to
99% conversion and up to 97% enantioselectivity using
0.5 mol% catalyst loading. The enantioselectivities ob-
tained using this novel catalytic system are clearly com-
patible to our previously reported results employing
pseudo-dipeptide-derived catalysts as well as to other
known catalytic systems. The major advantage with the
hydroxamic acid ligands is their simple preparation and
most importantly readily available and inexpensive start-
ing materials, which make this system highly competitive
in comparison to other ATH protocols.
Synlett 2007, No. 16, 2541–2544 © Thieme Stuttgart · New York

LETTER
Rhodium-Catalyzed Asymmetric Transfer Hydrogenation
2543
However, the outcome of the rhodium-catalyzed ATH
process using differently functionalized amino acids can-
not simply be explained by the NH acidity of the ligand.
Evidently, the coordination properties between ligands 1
and 2 are different, but at this point we are not able to pin-
point the reason for the observed enantioswitchable nature
of the catalysts formed from these compounds. We are
currently investigating this highly interesting feature and
hope to be able to present results on this in due time.
Table 2 Substrate Scope of the Rhodium-N-hydroxyl Amide
Catalyzed ATHa
OH
*
O
[RhCl2Cp*]2, 1b
R1
R2
i-PrONa, LiCl
i-PrOH
R1
R2
Entry R1
R2
Time
(h)
Convn ee
(%)b
82
93
98
77
99
54
90
85
84
(%, conf.)b
1
2
3
4
5
6
7
8
9
Ph
Me
2
2
2
2
2
2
2
2
2
97 (S)
92 (S)
96 (S)
94 (S)
95 (S)
81 (S)
86 (S)
90 (S)
87 (S)
In conclusion, we have introduced a new class of efficient
ligands for the rhodium-catalyzed ATH of aryl alkyl ke-
tones in 2-propanol. The ligands are easily accessed start-
ing from commercially available N-Boc-protected amino
acids. In addition we have demonstrated that enantioswit-
chable catalysts can be prepared from the same amino
acid depending on the choice of functionalities introduced
in the ligands. This feature can be most attractive in enan-
tioselective catalysis when only one enantiomer of the
precursor of the chiral ligand is available. We are current-
ly investigating the reason behind the enantioswitchable
nature observed using catalysts derived from differently
substituted amino acid ligands in the asymmetric transfer
hydrogenation of ketones.
4-BrC6H4
3-BrC6H4
2-BrC6H4
3-CF3C6H4
4-MeOC6H4
3-MeOC6H4
2-MeOC6H4
Ph
Me
Me
Me
Me
Me
Me
Me
Et
a Reaction conditions: substrate (1 equiv, 0.2 M in i-PrOH),
[{RhCl2Cp*}2] (0.25 mol%), ligand (0.55 mol%), LiCl (5 mol%) and
i-PrONa (5 mol% unless otherwise indicated), r.t.
b Conversion and enantioselectivity were determined by GLC (CP
Chirasil DEXCB).
Acknowledgment
This work was supported by the Swedish Research Council, the
Carl Trygger Foundation, and the Wenner-Gren Foundations (post-
doctoral fellowship to A.B.Z.).
The observed stereochemical outcome using rhodium cat-
alysts based on amino acid derived hydroxamic acids was
indeed unpredicted. As described above, in our previous
report on the use of thioamides derived from N-Boc-pro-
tected amino acids in the ATH of acetophenone we
achieved high conversion and enantioselectivity of (R)-1-
phenylethanol using ligand 2a (Figure 2).4 Using the dia-
stereomeric ligand 2b gave the same product absolute
configuration which show that the stereocenter present in
the amino acid apparently is directing the outcome of the
ATH reaction, presumably via the selective formation of
a dominant rhodium hydride. However, as presented in
this study, the use of the corresponding valine-derived hy-
droxamic acid ligand 1b resulted in a product with slightly
higher ee, albeit with opposite absolute configuration
(Table 1, entry 6).
References and Notes
(1) (a) Zassinovich, G.; Mestroni, G.; Gladiali, S. Chem. Rev.
1992, 92, 1051. (b) Ohkuma, T.; Noyori, R. In
Comprehensive Asymmetric Catalysis, Vol. 1; Jacobsen, E.
N.; Pfaltz, A.; Yamamoto, H., Eds.; Springer: Berlin, 1999,
199. (c) Blaser, H.-U.; Malan, C.; Pugin, B.; Spindler, F.;
Steiner, H.; Studer, M. Adv. Synth. Catal. 2003, 345, 103.
(d) Everaere, K.; Mortreux, A.; Carpentier, J.-F. Adv. Synth.
Catal. 2003, 345, 67. (e) Gladiali, S.; Alberico, E. In
Transition Metals for Organic Synthesis, Vol. 2; Beller, M.;
Bolm, C., Eds.; Wiley-VCH: Weinheim, 2004, 145.
(f) Zanotti-Gerosa, A.; Herns, W.; Groarke, M.; Hancock, F.
Platinum Met. Rev. 2005, 49, 158. (g) Ikariya, T.; Murata,
K.; Noyori, R. Org. Biomol. Chem. 2006, 4, 393.
(h) Noyori, R.; Hashiguchi, S. Acc. Chem. Res. 1997, 30, 97.
(i) Bianchini, C.; Glendenning, L. Chemtracts 1997, 10,
333. (j) Palmer, M. J.; Wills, M. Tetrahedron: Asymmetry
1999, 10, 2045. (k) Ohkuma, T.; Noyori, R. Comprehensive
Asymmetric Catalalysis, Suppl. 1; Jacobsen, E. N.; Pfaltz,
A.; Yamamoto, H., Eds.; Springer: New York, 2004, 43.
(l) Kitamura, M.; Noyori, R. In Ruthenium in Organic
Synthesis; Murahashi, S.-I., Ed.; Wiley-VCH: Weinheim,
2004, 3. (m) Gladiali, S.; Alberico, E. Chem. Soc. Rev. 2006,
35, 226. (n) Wu, X.; Xiao, J. Chem. Commun. 2007, 2449.
(2) For recent selected examples, see: (a) Brandt, P.; Roth, P.;
Andersson, P. G. J. Org. Chem. 2004, 69, 4885. (b) Kundu,
M. K.; Woggon, W.-D. Angew. Chem. Int. Ed. 2004, 43,
6731. (c) Xue, D.; Chen, Y.-C.; Cui, X.; Wang, Q.-W.; Zhu,
J.; Deng, J.-G. J. Org. Chem. 2005, 70, 3584. (d) Wu, X.;
Li, X.; King, F.; Xiao, J. Angew. Chem. Int. Ed. 2005, 44,
3407. (e) Hayes, A. M.; Morris, D. J.; Clarkson, G. J.; Wills,
M. J. Am. Chem. Soc. 2005, 127, 7318. (f) Baratta, W.;
S
Ph
S
Ph
N
H
N
H
NH
NH
2a
2b
Boc
Boc
88% conv, 95% ee (R)
78% conv, 95% ee (R)
Figure 2 Rhodium-catalyzed ATH of acetophenone using
thioamide ligands 2a and 2b
Consequently, the same N-Boc-protected amino acid can
by simple procedures be transformed into ligands, which
selectively give either S- or R-products in the ATH of aryl
alkyl ketones.
Synlett 2007, No. 16, 2541–2544 © Thieme Stuttgart · New York

2544
K. Ahlford et al.
LETTER
Chelucci, G.; Gladiali, S.; Siega, K.; Toniutti, M.; Zanette,
M.; Zangrando, E.; Rigo, P. Angew. Chem. Int. Ed. 2005, 44,
6214. (g) Enthaler, S.; Jackstell, R.; Hagemann, B.; Junge,
K.; Erre, G.; Beller, M. J. Organomet. Chem. 2006, 691,
4652. (h) Reetz, M. T.; Li, X. J. Am. Chem. Soc. 2006, 128,
1044.
(8) General Procedure for the Preparation of Hydroxamic
Acid Ligands 1a–d
To a solution of 2,4,6-trichloro-1,3,5-triazine (0.1 mmol) in
anhyd CH2Cl2 (8 mL) cooled to 0 °C, the following
components were added in the order they are written: Boc-
protected amino acid (3 mmol), NMM (6 mmol), DMAP
(0.3 mmol), and NH2OH·HCl (3 mmol). The reaction
mixture was stirred at r.t. for 14 h and thereafter filtered
through a plug of silica, using EtOAc as eluent. The residue
obtained after evaporation of the filtrate was chromatog-
raphed on silica (EtOAc–pentane, 10:1), followed by
recrystallization from acetone–pentane to give the
hydroxamic acids.
(3) (a) Pastor, I. M.; Västilä, P.; Adolfsson, H. Chem. Commun.
2002, 2046. (b) Pastor, I. M.; Västilä, P.; Adolfsson, H.
Chem. Eur. J. 2003, 9, 4031. (c) Bøgevig, A.; Pastor, I. M.;
Adolfsson, H. Chem. Eur. J. 2004, 10, 294. (d) Västilä, P.;
Zaitsev, A. B.; Wettergren, J.; Privalov, T.; Adolfsson, H.
Chem. Eur. J. 2006, 12, 3218.
(4) Zaitsev, A. B.; Adolfsson, H. Org. Lett. 2006, 8, 5129.
(5) For other examples of amino acid based ligands in the
transfer hydrogenation of ketones, see: (a) Ohta, T.;
Nakahara, S.; Shigemura, Y.; Hattori, K.; Furokawa, I.
Chem. Lett. 1998, 491. (b) Ohta, T.; Nakahara, S.;
Shigemura, Y.; Hattori, K.; Furokawa, I. Appl. Organomet.
Chem. 2001, 15, 699. (c) Carmona, D.; Lahoz, F. J.;
Atencio, R.; Oro, L. A.; Lamata, M. P.; Viguri, F.; San José,
E.; Vega, C.; Reyes, J.; Joó, F.; Kathó, Chem. Eur. J. 1999,
5, 1544. (d) Kathó, ; Carmona, D.; Viguri, F.; Remacha, C.
D.; Kovács, J.; Joó, F.; Oro, L. A. J. Organomet. Chem.
2000, 593-594, 209. (e) Carmona, D.; Lamata, M. P.;
Viguri, F.; Dobrinovich, I.; Lahoz, F. J.; Oro, L. A. Adv.
Synth. Catal. 2002, 344, 499. (f) Carmona, D.; Lamata, M.
P.; Oro, L. A. Eur. J. Inorg. Chem. 2002, 2239. (g) Faller,
J. W.; Lavoie, A. R. Organometallics 2001, 20, 5245.
(h) Rhyoo, H. Y.; Yoon, Y.-A.; Park, H.-J.; Chung, Y. K.
Tetrahedron Lett. 2001, 42, 5045. (i) Rhyoo, H. Y.; Park,
H.-J.; Chung, Y. K. Chem. Commun. 2001, 2064.
(j) Rhyoo, H. Y.; Park, H.-J.; Suh, W. H.; Chung, Y. K.
Tetrahedron Lett. 2002, 43, 269.
(6) Highly efficient vanadium catalysts containing ligands
based on hydroxamic acids were recently employed in the
asymmetric epoxidation of allylic alcohols, see: (a) Zhang,
W.; Basak, A.; Kosugi, Y.; Hoshino, Y.; Yamamoto, H.
Angew. Chem. Int. Ed. 2005, 44, 4389. For selected earlier
reports, see: (b) Michaelson, R. C.; Palermo, R. E.;
Sharpless, K. B. J. Am. Chem. Soc. 1977, 99, 1992.
(c) Murase, N.; Hoshino, Y.; Oishi, M.; Yamamoto, H.
J. Org. Chem. 1999, 64, 338. (d) Hoshino, Y.; Yamamoto,
H. J. Am. Chem. Soc. 2000, 122, 10452. (e) Bolm, C.;
Kühn, T. Synlett 2000, 899. (f) Bolm, C. Coord. Chem. Rev.
2003, 237, 245.
Compound 1a: yield 41%. 1H NMR (400 MHz, acetone-d6,
25 °C): d = 10.09 (s, 1 H), 8.22 (br s, 1 H), 6.06 (s, 1 H), 4.08
(q, J = 7.11 Hz, 1 H), 1.40 (s, 9 H), 1.29 (d, J = 7.11 Hz, 3
H). 13C NMR (100 MHz, acetone-d6, 25 °C): d = 170.0,
155.1, 78.3, 47.7, 27.5, 17.8.
Compound 1b: yield 25%. 1H NMR (400 MHz, acetone-d6,
25 °C ): d = 10.18 (br s, 1 H), 8.22 (br s, 1 H), 5.91 (d,
J = 8.16 Hz, 1 H), 3.75–3.85 (m, 1 H), 1.40 (s, 9 H), 0.89–
0.94 (m, 6 H). 13C NMR (100 MHz, acetone-d6, 25 °C ): d =
168.4, 155.4, 78.2, 57.5, 30.8, 27.5, 18.5, 17.6.
Compound 1c: yield 25%. 1H NMR (400 MHz, acetone-d6,
25 °C): d = 10.20 (br s, 1 H), 8.37 (br s, 1 H), 7.16–7.31 (m,
5 H), 6.12 (d, J = 7.32 Hz, 1 H), 4.23–4.36 (m, 1 H), 3.10
(dd, J = 13.71, 6.03 Hz, 1 H), 2.91 (dd, J = 13.71, 8.59 Hz, 1
H), 1.33 (s, 9 H). 13C NMR (100 MHz, acetone-d6, 25 °C):
d = 168.4, 155.1, 137.5, 129.2, 128.1, 126.3, 78.4, 53.6, 38.1,
27.5.
Compound 1d: yield 10%. 1H NMR (400 MHz, acetone-d6,
25 °C): d = 10.39 (br s, 1 H), 8.25 (br s, 1 H), 7.42–7.47 (m,
2 H), 7.26–7.37 (m, 3 H), 6.46 (d, J = 6.10 Hz, 1 H), 5.20 (d,
J = 6.10 Hz, 1 H), 1.39 (s, 9 H). 13C NMR (100 MHz,
acetone-d6, 25 °C): d = 167.3, 154.7, 138.9, 128.3, 127.6,
127.0, 78.6, 55.7, 27.5.
(9) For the original report on the importance of external base in
transition-metal-catalyzed transfer-hydrogenation reactions,
see: Chowdhury, R. L.; Bäckvall, J.-E. J. Chem. Soc., Chem.
Commun. 1991, 1063.
(10) General Procedure for the Transfer Hydrogenation of
Ketones Using Ligands 1a–d
[{RhCl2Cp*}2] (0.0025 mmol), ligand (0.0055 mmol), and
LiCl (0.05 mmol) were dried under vacuum in a dry Schlenk
tube for 15 min. Ketone (1 mmol), i-PrOH (4.5 mL), and a
0.01 M solution of i-PrONa in i-PrOH (0.5 mL, 5 mol%)
were added under nitrogen. The reaction mixture was stirred
at ambient temperature. Aliquots were taken after the
reaction times indicated in Tables 1 and 2 and were then
passed through a pad of silica with EtOAc as the eluent. The
resulting solutions were analyzed by GLC (CP Chirasil
DEXCB).
(7) Giacomelli, G.; Porcheddu, A.; Salaris, M. Org. Lett. 2003,
5, 2715.
(11) Turnover frequencies determined after 30 min reaction time.
Synlett 2007, No. 16, 2541–2544 © Thieme Stuttgart · New York