Four-coordinate dimethylgallium compounds vary in stability toward hydrolysis

Article abstract of DOI:10.1021/om980413x

Four-coordinate dimethylgallium complexes were prepared by the reaction of dimethylgallium hydroxide with bidentate Lewis bases and evaluated for their stability toward decomplexation in water, a key property in determining their potential usefulness as radiopharmaceuticals, particularly those targeted at specific receptor proteins. Model compounds from six structural classes, containing bidentate ligands with oxygen-oxygen, oxygen-nitrogen, and sulfur-nitrogen donor atoms, were prepared from the ligand and dimethylgallium hydroxide; they were characterized spectroscopically and, in two cases, by X-ray crystallography. The percentage decomplexation of the stable (CH₃)₂Ga⁺ unit from the Lewis base when exposed to 1000 equiv of water in acetone-d₆, determined by ¹H NMR, was used as a measure of the relative hydrolytic stability of each compound. Among the compounds we have investigated, the percent hydrolysis under these conditions ranged from 10% to 100%, the most hydrolytically stable compounds proving to be those based on an N-alkylsalicylaldimidate donor.

Full text of DOI:10.1021/om980413x

Organometallics 1998, 17, 4889-4896
4889
F ou r -Coor d in a te Dim eth ylga lliu m Com p ou n d s Va r y in
Sta bility tow a r d Hyd r olysis
Robert W. Chesnut,‡ Richard R. Cesati III,† Cathy S. Cutler,§ Sara L. Pluth,† and
J ohn A. Katzenellenbogen*,†
Department of Chemistry, University of Illinois, Urbana, Illinois 61801, Department of
Chemistry, Eastern Illinois University, Charleston, Illinois 61920, and Division of
Radiological Sciences, Mallinckrodt Institute of Radiology, Washington University Medical
School, St. Louis, Missouri 63110
Received May 22, 1998
Four-coordinate dimethylgallium complexes were prepared by the reaction of dimethyl-
gallium hydroxide with bidentate Lewis bases and evaluated for their stability toward
decomplexation in water, a key property in determining their potential usefulness as
radiopharmaceuticals, particularly those targeted at specific receptor proteins. Model
compounds from six structural classes, containing bidentate ligands with oxygen-oxygen,
oxygen-nitrogen, and sulfur-nitrogen donor atoms, were prepared from the ligand and
dimethylgallium hydroxide; they were characterized spectroscopically and, in two cases, by
X-ray crystallography. The percentage decomplexation of the stable (CH3)2Ga+ unit from
1
the Lewis base when exposed to 1000 equiv of water in acetone-d6, determined by H NMR,
was used as a measure of the relative hydrolytic stability of each compound. Among the
compounds we have investigated, the percent hydrolysis under these conditions ranged from
10% to 100%, the most hydrolytically stable compounds proving to be those based on an
N-alkylsalicylaldimidate donor.
In tr od u ction
So far, however, breast tumor imaging through the ER
has been accomplished only with a number of radio-
halogen-labeled estrogens. The availability of the halo-
gen radionuclides needed to produce these ER-binding
agents, however, is limited, either by their short half-
lives (e.g., for 18F, t1/2 ) 110 min) or by difficulties in
their production (e.g., production of 123I requires high-
energy cyclotrons). Because 66Ga and 67Ga have rela-
tively long half-lives (t1/2 ) 9.4 h and 3.3 days, respec-
tively) and 68Ga (t1/2 ) 68 min) is available from a long-
lived 68Ge (t1/2 ) 288 days) generator, replacing the
radiohalogen in these imaging agents with these gal-
lium radionuclides would make them more widely
available and might also simplify their preparation.
When gallium is used as the citrate salt of the Ga3+
cation, most in vivo transport and uptake of this cation
is mediated by iron-binding proteins, principally trans-
ferrin, because of the similar ionic radii of Ga3+ and
Radioisotopes of gallium, as the trication Ga3+, are
regularly used for medical imaging, and other gallium
compounds are being investigated for their potential as
radiopharmaceuticals.1-3 These radiodiagnostic agents
consist of ligands chelated to either 67Ga, 68Ga, or 66Ga
radioisotopes.4-6 Since the 66Ga and 68Ga isotopes are
positron-emitting nuclides with convenient half-lifes, we
became interested in incorporating gallium into estro-
gens for imaging of breast tumors by PET.
Breast tumors can be imaged by the binding of
appropriately labeled estrogens to the estrogen receptor
(ER), which is found in many tumors, and these images
provide useful prognostic information concerning cancer
stage and tumor responsiveness to hormone therapy.7-13
* Address correspondence to: J ohn A. Katzenellenbogen, Depart-
ment of Chemistry, University of Illinois, 461 Roger Adams Laboratory,
Box 37-5, 600 S. Mathews Ave., Urbana, IL 61801. Telephone: (217)
333-6310. Fax: (217) 333-7325. E-mail: jkatzene@uiuc.edu.
† University of Illinois.
Fe3+ 14
If undesirable for a given procedure, binding of
.
Ga3+ to transferrin can be prevented by the use of high-
affinity multidentate ligands, generally hexadentate
ones.15 An alternative form of gallium that has been
evaluated for radiopharmaceutical purposes is (CH3)2-
‡ Eastern Illinois University.
§ Mallinkrodt Institute of Radiology.
(1) Green, M. A.; Welch, M. J . Nucl. Med. Biol. 1989, 16, 435.
(2) Green, M. A. New Trends in Radiopharmaceutical Synthesis,
Quality Assurance, and Regulatory Control; Plenum Press: New York,
1991.
(9) Mintun, M. A.; Welch, M. J .; Siegel, B. A.; Mathias, C. J .;
Brodack, J . W.; McGuire, A. H.; Katzenellenbogen, J . A. Radiology
1988, 169, 45.
(10) McGuire, A. H.; Dehdashti, F.; Siegel, B. A.; Lyss, A. P.;
Brodack, J . W.; Mathias, C. J .; Mintun, M. A.; Katzenellenbogen, J .
A. J . Nucl. Med. 1991, 32, 1526.
(3) Green, M. A. Adv. Met. Med. 1993, 1, 75.
(4) Graham, M. C.; Pentlow, K. S. Med. Phys. 1997, 24, 317.
(5) Szelecse´nyi, F.; Boothe, T. E.; Tavano, E.; Plitnikas, M. E.;
Ta´rka´ny, F. Appl. Radiat. Isot. 1994, 45, 473.
(6) Goethals, P.; Coene, M.; Slegers, G.; Vogelaers, D.; Everaert, J .;
Lemahieu, I.; Colardyn, F.; Heyndrickx, G. R. Eur. J . Nucl. Med. 1990,
16, 237.
(7) Rijks, L. J . M.; Bakker, P. J . M.; Vantienhoven, G.; Noorduyn,
L. A.; Boer, G. J .; Rietbroek, R. C.; Taat, C. W.; J anssen, A. G. M.;
Veenhof, C. H. N.; Vanroyen, E. A. J . Clin. Oncol. 1997, 15, 2536.
(8) Rijks, L. J . M.; Sokole, E. B.; Stabin, M. G.; de Bruin, K.; J anssen,
A. G. M.; van Royen, E. A. Eur. J . Nucl. Med. 1998, 25, 40.
(11) Cummins, C. H. Steroids 1993, 58, 245.
(12) Pavlik, E. J .; Nelson, K.; Gallion, H. H.; van Nagell, J . R., J r.;
Donaldson, E. S.; Shih, W. J .; Spicer, J . A.; Preston, D. F.; Baranczuk,
R. J .; Kenady, D. E. Cancer Res. 1990, 50, 7799.
(13) Katzenellenbogen, J . A. In Estrogens, Progestins, and Their
Antagonists; Pavlik, E. J ., Ed.; Birkha¨user: Boston, 1996; p 197.
(14) Harris, W. R.; Pecoraro, V. L. Biochemistry 1983, 22, 292.
10.1021/om980413x CCC: $15.00 © 1998 American Chemical Society
Publication on Web 10/07/1998

4890 Organometallics, Vol. 17, No. 22, 1998
Chesnut et al.
F igu r e 1. Ligand precursors.
F igu r e 2. Structure of gallium complexes and percent
hydrolysis (after 12-24 h with 1000-fold excess of water
Ga+, a cation in which the metal-carbon bonds are
highly stable toward hydrolysis.1,16 The charge and size
of this cation apparently make it sufficiently dissimilar
to Fe3+ to prevent significant binding by plasma pro-
teins.16 Low affinity for plasma proteins, a tetracoor-
dinate (rather than hexacoordinate) geometry, and the
alkyl ligand environment of the metal suggest that
dimethylgallium compounds may be more adaptable to
the design of ligands for biological receptors than are
multidentate Ga3+ complexes.
The goal of this investigation was to prepare a
number of tetracoordinate compunds in which the
(CH3)2Ga moiety is complexed with a bidentate Lewis
base and to examine what structural factors affect the
stability of these complexes toward decomplexation of
the stable (CH3)2Ga moiety in the presence of water (i.e.,
hydrolysis) to determine whether complexes that are
stable to hydrolysis can be prepared. Developing stable
1
in acetone-d6 monitored by H NMR).
gallium-heteroatom bonds rather than stable gallium-
carbon bonds proved to be the challenge in this en-
deavor.
Resu lts a n d Discu ssion
P r ep a r a t ion of Liga n d s a n d Dim et h ylga lliu m
Com p ou n d s. The six classes of ligands and the six
classes of dimethylgallium compounds prepared from
them are shown in Figures 1 and 2, respectively. The
â-diketonate ligands (1a -d ) and their corresponding
complexes (7a -d ) represent species with two oxygen
donor atoms. The N-alkylaldimines (2a -e, 5 and 8a -
e, 11), N-arylaldimines (3a -g and 9a -g), and the
related N-arylbenzophenonimine analogues (4a ,b and
10a ,b) are all representatives of N,O-bidentate com-
plexes. Finally, the R-(2-pyridyl)acetophenones and
thioacetophenones (6a -d and 12a -d ) provide an al-
ternate N,O- and in one case an N,S-chelation geometry.
As is mentioned below, the design of several of these
(15) Sun, Y.; Andersen, C. J .; Pajeau, T. S.; Reichert, D. E.; Hancock,
R. D.; Motekaitis, R. J .; Martell, A. E.; Welch, M. J . J . Med. Chem.
1996, 39, 458.
(16) Coggin, D. K.; Mathias, C. J .; Green, M. L. Nucl. Med. Biol.
1994, 21, 283.

Four-Coordinate Dimethylgallium Compounds
Organometallics, Vol. 17, No. 22, 1998 4891
F igu r e 4. ORTEP plot of â-diketonate compound 7c with
thermal ellipsoids at 35%.
F igu r e 3. Classes of nonsteroidal estrogens related to
dimethyl gallium complexes.
Ta ble 1. P r ep a r a tion of Ga lliu m Com p lexes
ligand
(CH3)2GaOH
excess
(%)
mass
mass
(mg)
yield
(%)
no.
(mg)
mmol
mmol
7c
7d
8a
8
8d
8e
9
9b
9c
9d
9e
9f
9g
10a
10b
11
12a
12b
12c
12d
81.3
23.2
84.6
81.2
131.1
94.8
61.4
71.7
72.7
81.2
81.8
64.0
40.0
71.6
122.0
76.1
115.0
104.0
200.0
50.0
0.29
0.09
0.43
0.38
0.57
0.33
0.32
0.37
0.33
0.27
0.36
0.31
0.18
0.23
0.38
0.34
0.58
0.49
0.94
0.23
35.4
11.1
52.7
46.7
115.3
63.2
39.4
45.9
40.0
32.4
45.2
38.0
22.0
28.8
71.4
42.2
73.0
61.0
117.0
29.0
0.30
0.10
0.45
0.40
0.99
0.54
0.34
0.39
0.34
0.28
0.39
0.33
0.19
0.25
0.62
0.36
0.62
0.52
0.99
0.25
6
11
6
84
74
97
100
97
85
100
95
92
96
86
100
95
85
76
78
97
92
91
89
F igu r e 5. ORTEP plot of R-(2-pyridyl)acetophenone 12a
with thermal ellipsoids at 35%.
5
70
63
6
6
4
5
8
5
4
6
60
6
7
6
12a (Figure 5), confirm a four-coordinate environment
about the metal and bond lengths to gallium that are
similar to those in other known complexes; details are
given in Tables 3 and 4.17 Reflecting the electropositive
character of the metal center, NMR spectra of the
compounds consistently show the dimethylgallium hy-
drogens and carbons slightly upfield of TMS. There is,
however, remarkably little difference between NMR
spectra of the gallium complexes and those of the free
ligand and dimethylgallium hydroxide itself. The 1H
NMR resonances of the methyl groups in dimethylgal-
lium hydroxide (δ -0.48) appear at somewhat lower
field in the complexes (ca. δ -0.46), the shift in the
resonance position of these methyl groups upon hy-
drolysis of the complex being generally 0.03 ppm down-
field. Similarly, the resonances for most of the aromatic
and vinylic protons in these systems appear at some-
what higher field in the complexes than in the free
ligands; again, the shift is small, ca. 0.05-0.15 ppm.
Mass spectra show a peak for the molecular ion (M+)
or, more commonly, the mass peak resulting from
monodemethylation at the gallium center. In either
case, the mass peaks corresponding to the two stable
isotopes of gallium (69Ga and 71Ga) show the expected
relative intensities of 3:2.
5
9
Sch em e 1. Syn th esis of
r-(2-P yr id yl)a cetop h en on es
gallium compounds was inspired by the structure of
certain nonsteroidal ligands for the estrogen receptor
(Figure 3). The synthesis of the ligands followed
literature precedent or was accomplished using well-
precedented methods for aromatic substitution and
imine formation reactions. Procedures for the synthesis
of new compounds are presented in detail in the
Experimental Section and in Scheme 1 and will not be
discussed further. The reaction scale and yield of
complex formation are given in Table 1.
By all structural comparisons, the dimethylgallium
complexes 7-12 are similar to reported compounds.
Crystal structures of two of the compounds, the â-dike-
tonate 7c (Figure 4) and the R-(2-pyridyl)acetophenone
Hyd r olysis Stu d ies. There are reports that the
stability of trialkylgallium complexes toward hydrolysis
can be improved by intramolecular chelation.18,19 There
are also qualitative reports of the hydrolytic stability
of dimethylgallium compounds of the type we studied.
For example, dimethylgallium complexes of substituted
salicylideneimine and dimethylaminomethyl-3-pyridol-
(17) Schumann, H.; Hartmann, U.; Dietrich, A.; Pickardt, J . Angew.
Chem., Int. Ed. Engl. 1988, 27, 1077.
(18) Schumann, H.; Hartmann, U.; Wasserman, W. Polyhedron
1990, 9, 353.
(19) Schumann, H.; Hartmann, U.; Wassermann, W.; Dietrich, A.;
Gorlitz, F.; Pohl, L.; Hostalek, M. Chem. Ber. 1990, 123, 2093.

4892 Organometallics, Vol. 17, No. 22, 1998
Chesnut et al.
Ta ble 2. Cr ysta llogr a p h ic Deta ils
7c
complex
12a
empirical formula
fw
C
19H21GaO4
C15H16GaNO
296.01
383.08
temp
198(2) K
198(2) K
wavelength
cryst syst
space group
unit cell dimens
0.710 73 Å
orthorhombic
Pnma
0.710 73 Å
monoclinic
P21/n
a ) 7.915 Å
a ) 9.3261(2) Å
b ) 11.6182(2) Å
c ) 12.9708(3) Å
1369.14(5) Å3
4
b ) 21.0640(4) Å
c ) 10.5510(2) Å
1759.08(5) Å3
4
volume
Z
density, calcd
abs coeff
no. of indep reflns
refinement method
no. of data/restraints/params
goodness-of-fit (GoF) on F2
final R indices (I > 2σ(I))
R indices (all data)
1.446 g cm-3
1.583 mm-1
1.436 g cm-3
1.997 mm-1
8701 (R(int) ) 0.0533)
10688 (R(int) ) 0.0384)
full matrix least-squares on F2
2213/0/115
full matrix least-squares on F2
3297/0/163
1.101
1.124
R1 ) 0.0518, wR2 ) 0.1420
R1 ) 0.0635, wR2 ) 0.1536
R1 ) 0.0334, wR2 ) 0.0652
R1 ) 0.0496, wR2 ) 0.0744
Ta ble 3. Selected Bon d Dista n ces a n d An gles for
heteroatom bond in our dimethylgallium complexes,
rather than from hydrolysis of a Ga-carbon bond in a
trialkyl gallium compound.
â-Dik eton a te Com p lex 7c
Bond Distances (Å)
1
Ga(1)-O(1)
Ga(1)-C(10)
Ga(1)-C(11)
1.921(2)
1.959(4)
2.004(3)
Because it can be monitored by H NMR, the hydroly-
sis process can be conveniently followed. Hydrolysis
appears to be relatively rapid, and in the cases where
the kinetics were followed in some detail, it reached a
limiting value within a few hours. So, the extent of
hydrolysis was routinely measured after 12-24 h, at
which point it had reached an equilibrium value.
Because the hydrolysis assay was done at a fixed
concentration of dimethylgallium complex (200 mM, see
Experimental Section) and a fixed, 1000-fold excess of
water, we are measuring the relative thermodynamic
stabilities of these complexes. The results of the hy-
drolysis studies on the complexes 7-12 are summarized
in Figure 2.
â-Dik eton a te Com p ou n d s 7a -7c. Complexes of
this structure were inspired by certain nonsteroidal
ligands for the estrogen receptor of the benzestrol class
(see Figure 3). Hydrolysis studies on these complexes
demonstrated how chelate structure affected hydrolytic
stability (see Figure 2). The fact that the published
synthesis of 7a involves an extraction of excess acetyl-
acetonate with a water/diethyl ether system makes
(CH3)2Ga(acac) appear stable to hydrolysis. By contrast,
our measurements show that, in acetone, (CH3)2Ga-
(acac) is more than 50% hydrolyzed by 1000 equiv of
water. These observations can be reconciled by the
equilibrium
Bond Angles (deg)
O(1)-Ga(1)-O(1A)
O(1)-Ga(1)-C(10)
O(1)-Ga(1)-C(11)
C(10)-Ga(1)-C(11)
C(8)-O(1)-Ga(1)
93.89(14)
108.88(12)
106.43(10)
127.2(2)
124.4(2)
Ta ble 4. Selected Bon d Dista n ces a n d An gles for
r-(2-P yr id yl)Acetop h en on e Com p lex 12a
Bond Distances (Å)
Ga(1)-O(1)
Ga(1)-C(14)
1.899(2)
1.958(3)
Ga(1)-C(15)
Ga(1)-N(1)
1.960(3)
2.027(2)
Bond Angles (deg)
108.99(10) C(15)-Ga(1)-N(1) 106.40(10)
O(1)-Ga(1)-C(14)
O(1)-Ga(1)-C(14)
109.78(10) C(1)-N(1)-Ga(1)
117.4(2)
123.3(2)
124.0(2)
C(14)-Ga(1)-C(15) 125.71(11) C(5)-N(1)-Ga(1)
O(1)-Ga(1)-N(1)
C(14)-Ga(1)-N(1)
94.10(7)
C(7)-O(1)-Ga(1)
107.13(10)
ato ligands are described as “air and moisture sensi-
tive”20,21 or “air stable”, respectively.20,21
In some quantitative hydrolysis studies of trialkyl-
gallium or dialkylphosphidogallium compounds, the
reaction of the complex with small amounts (usually 1
equiv) of water was studied. Aggregated dialkylgallium
hydroxides are produced by these reactions.22-24 Our
experiments were similar, in that [(CH3)2GaO]n was the
observed product of hydrolysis; however, our experi-
ments were designed to allow quantitative comparisons
of hydrolytic stability in screening for potential radio-
pharmaceuticals. Thus, in contrast to this earlier work,
we added a relatively large quantity of water (1000
equiv) to the gallium compound. Also, the hydroxide
product resulted from addition of water across the Ga-
Preferential solubility of (CH3)2Ga(acac) in ether and
removal of water by bulk partitioning and chemical
drying would explain how the unstable (CH3)2Ga(acac)
can be prepared using an aqueous workup. A second
implication of our results is that the in vivo distribution
of [67Ga](CH3)2Ga(acac) is probably largely controlled by
its hydrolysis product, [67Ga](CH3)2GaOH.
(20) Chong, K. S.; Rettig, S. J .; Storr, A.; Trotter, J . Can. J . Chem.
1977, 55, 2540.
(21) Onyiriuka, E. C.; Rettig, S. J .; Storr, A.; Trotter, J . Can. J .
Chem. 1987, 65, 782.
(22) Power, M. B.; Cleaver, W. M.; Apblett, A. W.; Barron, A. A.
Polyhedron 1992, 11, 477.
(23) Naiini, A. A.; Young, V.; Han, Y.; Akinc, M.; Verkade, J . G.
Inorg. Chem. 1993, 32, 3781.
(24) Atwood, D. A.; Cowley, A. H.; Harris, P. R.; J ones, R. A.;
Koschmieder, S. U.; Nunn, C. M. Organometallics 1993, 12, 24.
With identical heteroatom donors, the â-diketonate
complexes 7 are a simple model for probing the differ-
ences in hydrolytic stability that can result from struc-
tural changes distal to the metal (see Figure 2). Elec-
tronic influences are most obvious in explaining the
relative stabilities of 7a vs 7b. Replacing the methyl

Four-Coordinate Dimethylgallium Compounds
Organometallics, Vol. 17, No. 22, 1998 4893
group in 7a with the more electron-withdrawing phenyl
groups in 7b lowers the basicity and thereby reduces
the reactivity of the complexed ligand toward water.
Addition of electron-donating methoxy groups in 7c
partially reverses the decrease in basicity and, therefore,
increases hydrolytic stability.
gens (see Figure 3). The marked hydrolytic instability
of 10b is significant by comparison with that of 8b.
Replacement of the aldimine hydrogen (of 8b) with a
p-methoxyphenyl group (as in 10b) would be expected
to reduce the basicity of the imine nitrogen and,
consequently, the donor character of the ligand toward
(CH3)2Ga+, increasing its stability, as was observed.
Steric crowding in this complex is also expected to twist
this phenyl substitutent largely out of conjugation with
the aldimine.
N-Ar yla ld im in e Com p ou n d s (8a -e). The struc-
ture of this class of gallium compounds was inspired by
the structure of the nonsteroidal estrogens of the
stilbestrol class (see Figure 3). These complexes were
the first system in which we separately tuned the
basicities of the two different donor atoms of the ligand.
Specifically, we designed a series to increase the basicity
of the N-donor and decrease the basicity of the O-donor
in an attempt to maximize ligand affinity for (CH3)2-
Ga+ and minimize reactivity of the dimethylgallium
complex toward water. Implementing this trend with
substituents generally increased the hydrolytic stability
of the compounds (8a -d ) as desired (see Figure 2). In
this series, we used hydroxyl substituents to alter donor
atom basicity because estrogenicity, a target property,
is often engendered by the phenolic OH functionality.25
The fact that the most acidic oxygen donor ligand (8e,
bearing a nitro group) did not have the greatest hydro-
lytic stability suggests that a maximum desirable acidity
can be exceeded and that in this ligand the σ-donor
character of the oxygen donor toward (CH3)2Ga+ was
diminished.
N-Alk yla ld im in e Com p ou n d s (9a -g). The hydro-
lytic behavior of these complexes is significant because
of the emergence of a new parameter influencing
hydrolytic stability. The large hydrolytic labilizing
effect of substituting a tert-butyl group (9a ) for an
n-butyl group (9b) is evidence that the bulk of a
substituent on the imine nitrogen can compromise
hydrolytic stability (see Figure 2). This effect may
reflect the relative thermodynamic stabilities of the two
complexes. The length of an n-alkyl chain, however, had
little effect on the reactivity of these compounds (9b-
e) toward water. Removal of the meta hydroxyl group
(9f) afforded the most hydrolytically stable member of
the series. Addition of an aryl (9e) or ester (9g)
functionality at the ω-carbon did not significantly affect
the hydrolytic stability of these complexes.
Com p ou n d 11. Complex 11 could be considered in
the class of N-alkylaldimine complexes (9), but was
studied separately, for comparative purposes. The low
hydrolytic stability of complex 11 is consistent with two
earlier observations: (a) the steric crowding caused by
a N-(tert-butyl) group (as in complex 9a ) and (b) the
(apparently) low donor character of the p-nitrophenoxide
ligand (as in complex 8e). Improvement in stability
through a better match of acid/base softness was probed
by substituting sulfur for oxygen. We found, however,
that the S-donor complex could not be formed from the
free thiophenol (5b) and (CH3)2GaOH, even in refluxing
THF. The sulfur atom may be sufficiently larger than
oxygen to cause significant strain in the incipient six-
member ring that includes the (CH3)2Ga+ moiety.
r-P yr id yla cetop h en on e a n d Th ioa cetop h en on e
Com p ou n d s (12a -d ). This system, which provides an
alternate geometry for bidentate complexation of the
dimethylgallium cation, was systematically modified
using hydroxyl groups to probe the effect of increasing
the basicity of the N-donor and decreasing the basicity
of the O-donor (as was done in the series 8a -d ), in an
attempt to maximize ligand affinity for (CH3)2Ga+ and
minimize the reactivity of the complex toward water.
However, in contrast to the results in the 8a -d series,
a marked decrease in hydrolytic stability was observed
with increasing basicity of the O-donor and decreasing
basicity of the N-donor (12b and 12c vs 12a ). Substi-
tuting the O-donor for an S-donor increased the acidity
of the ligand and hence its σ-donor ability toward (CH3)2-
Ga+, affording a somewhat greater stability to this
complex (12d ) than to its oxygen congener (12a ). The
differences in the response of the hydrolytic stability of
the dimethylgallium compounds in the series 8a -d and
12a -c to hydroxyl substitution could be due to the
different paths of conjugation between the hydroxyl
groups and the N-, and O-donor atoms in the two
complexes, as their structures are rather different.
Approximately the same range of hydrolytic stability
is present across series 8 and 9, but N-(n-alkyl) groups
in complexes 9 stabilize the complexes more than does
an unsubstituted N-phenyl group in complexes 8. This
difference probably parallels basicity differences at the
imine nitrogen. Why a meta OH group should have
opposite results on hydrolytic stability in the two series
is not clear, however: 8c (with a m-OH) is more stable
than 8b, whereas 9a (with a m-OH) is less stable than
9f. If these complexes are being stabilized against
hydrolysis by formation of micellar structures in which
the Ga-O bonds are being sequestered away from water
where they are selectively protected from hydrolysis,
then micellar destabilization by a polar substituent on
the aromatic ring would be understandable.
Con clu sion
We have shown that four-coordinate dimethylgallium
compounds exhibit marked differences in their hydro-
lytic stability in the presence of 1000-fold excess of
water. In our model systems, stability of substituted
salicylideneimine complexes was favored by basicity and
steric accessibility of the imine nitrogen. Stability
decreased when the acidity of the phenolic oxygen was
increased by a p-nitro group. Sulfur donor complexes
showed only marginal improvement in stability relative
to O-donor analogues.
N-Ar yl-Hydr oxyben zoph en on im in e Com pou n ds
(10a a n d 10b). These compounds were prepared
because of their similarity to centchromene-type estro-
Future efforts to develop hydrolytically stable dimeth-
ylgallium complexes as radiopharmaceuticals should
concentrate on ligands different from those we have
described. Alternatives include five-coordinate gallium
(25) Anstead, G. M.; Carlson, K. E.; Katzenellenbogen, J . A. Steroids
1997, 62, 268.

4894 Organometallics, Vol. 17, No. 22, 1998
Chesnut et al.
7.3, 7.3, 1.0 Hz), -0.26 (s, 6H). 13C NMR (acetone-d6): δ 169.8,
compounds and dialkyl compounds in which the phar-
macophore is included in the alkyl group(s). In this
regard, it is of note that none of the dimethylgallium
complexes underwent hydrolytic scission of the gal-
lium-alkyl bond.
168.2, 148.0, 137.8, 137.4, 130.6, 130.3, 128.5, 123.0, 122.6,
+
122.0, 119.7, 116.9, -6.4. LRMS (EI): M - CH3 ) 280.
HRMS (EI) M+ Calcd for C15H16NOGa: 295.0488. Found:
295.0487.
K2N,O-2-[(4-H yd r oxyp h en ylim in o)m et h yl]p h en ola t o-
d im eth ylga lliu m (8b). 1H NMR (acetone-d6): δ 8.78 (s, 1H),
8.52 (s, 1H), 7.43 (d, 1H, J ) 7.8 Hz), 7.38 (dd, 1H, J ) 6.8,
6.8 Hz), 7.29 (d, 2H, J ) 8.8 Hz), 6.98 (d, 2H, J ) 8.8 Hz),
6.77 (d, 1H, J ) 8.5 Hz), 6.71 (dd, 1H, J ) 7.4, 7.4 Hz). 13C
NMR (acetone-d6): δ 167.3, 167.0, 157.3, 139.4, 136.5, 136.4,
Exp er im en ta l Section
Liga n d Syn th esis. Compounds 1a , 1b, and 2a were
purchased from Sigma-Aldrich. Published procedures were
employed to synthesize 1c,26,27 2b,28 2c,29 and 5b.30 Other
ligand precursors (1d , 2d e, 3a -g, 4a ,b, 6a -d , 21c, 22c) were
synthesized by the methods described in the Supporting
Information.
+
123.0, 122.1, 119.1, 116.4, 116.1, -7.1. LRMS (EI): M - CH3
) 296. HRMS (EI) M - CH3+ Calcd for C13H11NO2: 296.0202.
Found: 296.0204.
K2N,O-4-[(4-Hyd r oxyp h en ylim in o)m eth yl]ben zen e-1,3-
d iola tod im eth ylga lliu m (8c). 1H NMR (1:4 CD3OD/acetone-
d6): δ 8.31 (s, 1H), 7.24 (d, 1H, J ) 8.6 Hz), 7.17 (d, 2H, J )
8.9 Hz), 6.89 (d, 1H, J ) 9.0 Hz), 6.24 (dd, 1H, J ) 10.0, 2.0
Hz), 6.17 (d, 1H, J ) 2.7 Hz), -0.26 (s, 6H). 13C NMR (acetone-
d6): δ 169.9, 166.2, 157.4, 140.4, 139.2, 123.5, 117.0, 113.9,
107.4, 106.9, -6.4. LRMS (EI): M - CH3+ ) 312. HRMS (EI)
M+ Calcd for C15H16NO3Ga: 327.0386. Found: 327.0382.
K2N,O-4-[(4-N,N-Dim eth yla m in op h en ylim in o)m eth yl]-
ben zen e-1,3-d iola tod im eth ylga lliu m (8d ). 1H NMR (ace-
tone-d6): δ 8.33 (s, 1H), 7.27 (d, 1H, J ) 8.5 Hz), 7.22 (d, 2H,
J ) 8.8 Hz), 6.82 (d, 2H, J ) 8.8 Hz), 6.28 (dd, 1H, J ) 8.5,
2.2 Hz), 6.20 (s, 1H), 3.01 (s, 6H), -0.34 (s, 6H). 13C NMR
(acetone-d6): δ 169.1, 163.9, 149.9, 138.2, 136.6, 122.1, 113.3,
112.9, 106.5, 106.2, 39.8, -7.1. LRMS (EI): M - CH3+ ) 339.
Anal. Calcd for C17H21N2O2Ga: C, 57.20; H, 5.96; N, 7.89.
Found: C, 57.49; H, 6.14; N, 7.59.
K2N,O-2-[(4-N,N-Dim eth yla m in op h en ylim in o)m eth yl]-
4-n itr op h en ola tod im eth ylga lliu m (8e). 1H NMR (acetone-
d6) δ 8.81 (s, 1H), 8.53 (d, 1H, J ) 2.9 Hz), 8.20 (dd, 1H, J )
9.2, 2.9 Hz), 7.41 (d, 1H, J ) 9.0 Hz), 6.87 (d, 1H, J ) 9.0 Hz),
6.87 (d, 1H, J ) 9.0 Hz), 3.04 (s, 6H), -0.23 (s, 6H). 13C NMR
(acetone-d6) δ 123.4, 123.2, 113.4, 40.3, -6.1. LRMS (EI)
M-CH3+ ) 368. HRMS (EI) M+ Calcd for C17H21N3O3Ga:
387.0760. Found: 387.0761.
Ga lliu m Com p lexes. Gen er a l P r oced u r e for F or m a -
tion of Ga lliu m Com p lexes. Dimethylgallium hydroxide
and complex 7a were prepared by a published procedure.31,32
Except where otherwise noted, the remaining gallium com-
plexes were prepared as follows. A 2 mL flask was charged
with 70-125 mg of ligand precursor, a 5-8% molar excess (of
1:1 stoichiometry) of (CH3)2GaOH, and 0.5-0.7 mL of THF.
The resulting solution was stirred 8-12 h; then MgSO4 was
added, and the mixture was stirred an additional 10 min.
Filtration through Celite afforded a clear filtrate from which
solvent was removed in vacuo to afford a solid. Since the
complexes proved unstable on alumina, silica, and C18 silica,
they were washed with three 5 mL portions of boiling
petroleum ether to remove excess (CH3)2GaOH and dried in
vacuo. Compounds 9f and 9g were soluble in petroleum ether;
they were characterized with 4-5% excess (CH3)2GaOH
present.
K2O,O-Diben zoylm eth a n ola tod im eth ylga lliu m (7b). To
0.53 g (5.21 mmol) of (CH3)2GaOH dissolved in 5 mL of (1:1:1)
benzene/hexane/Et2O was added a solution of 1.17 g (5.21
mmol) of dibenzoylmethane in 7 mL of benzene. The solution
was stirred for 12 h at 22 °C. Removal of solvent in vacuo
afforded a solid that was subsequently recrystallized from
hexane to afford yellow needles. Yield: 0.38 g (22%). Ad-
ditional product was obtained by concentration of the mother
liquor. 1H NMR (acetone-d6): δ 8.14 (d, 4H, J ) 7.1 Hz), 7.62-
7.51 (m, 6H), 7.06 (s, 1H), -0.16 (s, 6H). 13C NMR (acetone-
d6): δ 186.8, 138.5, 133.5, 129.5, 129.4, 129.2, 128.6, 128.1,
K2NO-4-ter t-Bu t ylim in om et h ylb en zen e-1,3-d iola t od i-
m eth ylga lliu m (9a ). 1H NMR (acetone-d6): δ 8.34 (s, 1H),
7.17 (d, 1H, J ) 8.6 Hz), 6.21 (dd, 1H, J ) 8.5, 2.0 Hz), 6.14
(d, 1H, J ) 1.8 Hz), 1.43 (s, 9H), -0.34 (s, 6H). 13C NMR
(acetone-d6): δ 172.8, 169.2, 166.3, 138.7, 137.7, 113.3, 106.6,
+
123.1, -6.8. LRMS (EI): M - CH3 ) 307. Anal. Calcd for
C
17H17O2Ga: C, 63.21; H, 5.30. Found: C, 62.84; H, 5.11.
+
60.1, -4.0. LRMS (EI): M - CH3 ) 276. HRMS (EI) M+
K2O,O-1,3-Bis(4-m eth oxyp h en yl)p r op a n e-1,3-d ion ola -
Calcd for C13H20NO2Ga: 291.0750. Found: 291.0751.
tod im eth ylga lliu m (7c). 1H NMR (CDCl3): δ 7.95 (d, 4H, J
) 9.4 Hz), 6.92 (d, 4H, J ) 8.8 Hz), 6.64 (s, 1H), 3.88 (s, 6H),
-0.17 (s, 6H). 13C NMR (CDCl3): δ 184.5, 163.3, 130.8, 130.6,
K2N ,O-4-n -Bu t ylim in om e t h ylb e n ze n e -1,3-d iola t od i-
m eth ylga lliu m (9b). 1H NMR (acetone-d6): δ 8.21 (s, 1H),
7.14 (d, 1H, J ) 8.5 Hz), 6.25 (dd, 1H, J ) 8.5, 2.2 Hz), 6.19
(d, 1H, J ) 2.0 Hz), 3.59 (t, 2H, J ) 7.6 Hz), 1.70 (tt, 2H, J )
7.3, 7.3 Hz), 1.41 (tt, 2H, J ) 7.5, 7.5 Hz), 0.98 (t, 3H, J ) 4.2
Hz), -0.37 (s, 6H). 13C NMR (acetone-d6): δ 169.5, 169.1,
138.0, 113.0, 106.9, 106.6, 58.4, 32.9, 20.5, 13.9, -7.0. LRMS
+
129.9, 114.3, 114.0, 92.5, 55.7, -6.5. LRMS (EI): M - CH3
) 367. Anal. Calcd for C19H21O4Ga: C, 59.57; H, 5.33.
Found: C, 59.66; H, 5.40.
K2O,O-1,3-Bis(4-h yd r oxyp h en yl)p r op a n e-1,3-d ion ola to-
d im eth ylga lliu m (7d ). 1H NMR (acetone-d6): δ 9.18 (s, 2H),
8.04 (d, 4H, J ) 8.8 Hz), 6.94 (d, 4H, J ) 8.8 Hz), 6.91 (s, 1H),
-0.25 (s, 6H). 13C NMR (acetone-d6): δ 184.9, 162.5, 130.9,
130.0, 123.2, 116.1, 92.2, -7.0. LRMS (EI): M - CH3+ ) 339.
Anal. Calcd for C17H17O4Ga: C, 57.51; H, 4.82. Found: C,
57.16; H, 4.40.
(EI): M - CH3 ) 276. HRMS (EI) M+ calcd for C13H20NO2-
+
Ga: 291.0750. Found: 291.0748.
K2N,O-4-n -H exylim in om et h ylb en zen e-1,3-d iola t od i-
m eth ylga lliu m (9c). 1H NMR (acetone-d6): δ 8.18 (s, 1H),
7.10 (d, 1H, J ) 8.6 Hz), 6.21 (dd, 1H, J ) 8.4, 2.4 Hz), 6.16
(d, 1H, J ) 2.4 Hz), 3.56 (t, 2H, J ) 6.4 Hz), 1.65 (tt, 2H, J )
7.1, 7.1 Hz), 1.32-1.34 (m 6H), 0.89 (t, 3H, J ) 7.1 Hz), -0.34
(s, 6H). 13C NMR (acetone-d6): δ 169.5, 169.1, 165.5, 138.0,
113.0, 107.0, 106.7, 58.7, 32.1, 30.9, 27.0, 23.1, 14.2, -7.0.
K2N,O-2-P h e n ylim in om e t h ylp h e n ola t od im e t h ylga l-
liu m (8a ). 1H NMR (acetone-d6): δ 8.60 (s, 1H), 7.58 (dd, 1H,
J ) 8.4, 8.4 Hz), 6.83 (d, 1H, J ) 8.5 Hz), 6.76 (ddd, 1H, J )
+
+
LRMS (EI): M - CH3 ) 304. HRMS (EI) M - CH3 calcd
for C14H21NO2Ga: 304.0828. Found: 304.0827.
(26) Adams, J . T.; Hauser, C. R. J . Am. Chem. Soc. 1944, 66, 1220.
(27) Levine, R.; Conroy, J . A.; Adams, J . T.; Hauser, C. R. J . Am.
Chem. Soc. 1945, 67, 1510.
(28) Haegle, E. Chem. Ber. 1892, 25, 2753.
(29) Helfrich, B.; Mitrowsky, A. Chem. Ber. 1952, 85, 1.
(30) Korobov, M. S.; Minkin, V. I. J . Org. Chem. USSR 1975, 11,
826.
(31) Coates, G. E.; Hayter, R. G. J . Chem. Soc. 1953, 2519.
(32) Tobias, R. S.; Sprague, M. J .; Glass, G. E. Inorg. Chem. 1968,
7, 1714.
K2N,O-4-Dod ecylim in om et h ylb en zen e-1,3-d iola t od i-
m eth ylga lliu m (9d ). 1H NMR (acetone-d6): δ 8.16 (s, 1H),
7.10 (d, 1H, J ) 8.5 Hz), 6.21 (dd, 1H, J ) 8.4, 2.4 Hz), 6.17
(d, 1H, J ) 2.4 Hz), 3.55 (t, 2H, J ) 7.6 Hz), 1.68 (m, 2H),
1.35-1.20 (m, 18H), 0.88 (t, 3H, J ) 6.4 Hz), -0.40 (s, 6H).
13C NMR (acetone-d6): δ 169.4, 169.0, 165.4, 137.9, 113.0,
107.0, 106.7, 58.7, 32.6, 30.8, 30.3, 30.1, 27.3, 23.3, 14.3, -6.9.

Four-Coordinate Dimethylgallium Compounds
Organometallics, Vol. 17, No. 22, 1998 4895
LRMS (EI): M - CH3+ ) 388. HRMS (EI) M+ calcd for C21H36
-
138.5, 130.9, 127.7, 122.8, 117.4, 114.8, 92.0, -7.95. MS (EI,
70 eV): m/ z (relative intensity) 313 (M+, 7), 311 (M+, 9), 298
(68), 296 (100), 282 (7), 280 (11), 179 (15), 177 (24), 149 (10),
NO2Ga: 403.2002. Found: 403.2003.
K2N ,O-4-Be n zylim in om e t h ylb e n ze n e -1,3-d iola t od i-
m eth ylga lliu m (9e). 1H NMR (acetone-d6): δ 8.39 (s, 1H),
7.39-7.34 (m, 5H), 7.17 (d, 1H, J ) 8.5 Hz), 6.24 (d, 1H, J )
8.5 Hz), 6.15 (d, 1H, J ) 2.0 Hz), 4.74 (s, 2H), -0.67 (s, 6H).
13C NMR (acetone-d6): δ 169.8, 169.4, 138.3, 137.1, 129.8,
129.5, 128.9, 113.1, 106.9, 106.8, 62.4, -7.2. LRMS (EI): M
- CH3+ ) 310. HRMS (EI) calcd for C15H15NO2Ga: 310.0359.
Found: 310.0358.
148 (14), 121 (20), 71 (42), 69 (65). Anal. Calcd for C15H16
-
GaNO2‚0.5H2O: C, 56.12; H, 5.34; N, 4.36. Found: C, 56.56;
H, 5.32; N, 5.32.
K2N,O-2-(5-Hyd r oxy-2-p yr id yl)-1-p h en yl-1-eth en ola to-
d im eth ylga lliu m (12c): Yellow solid. 1H NMR ((CD3)2CO,
500 MHz): δ 9.08 (bs, 1H), 7.86 (dd, 2H, J ) 8.14, 1.57 Hz),
7.72 (d, 1H, J ) 2.80 Hz), 7.48 (dd, 1H, J ) 8.88, 2.77 Hz),
7.34 (m, 3H), 7.29 (d, 1H, J ) 8.96 Hz), 6.11 (s, 1H), -0.29 (s,
6H). 13C NMR (CDCl3, 125 MHz): δ 165.6, 151.6, 150.7, 141.2,
130.8, 129.7, 129.4, 128.7, 126.7, 125.7, 93.6, -7.80. MS (EI,
70 eV): m/ z (relative intensity) 313 (M+, 5), 311 (M+, 6), 298
(60), 296 (88), 282 (10), 280 (121), 212 (4), 184 (10), 163 (13),
161 (18), 149 (4), 148 (5), 105 (31), 77 (34), 71 (65), 69 (100).
Anal. Calcd for C15H16GaNO2: C, 57.74; H, 5.17; N, 4.49.
Found: C, 57.59; H, 4.94; N, 4.27.
K2NO-2-n -Hexylim in om et h ylp h en ola t od im et h ylga lli-
u m (9f). 1H NMR (acetone-d6): δ 8.37 (s, 1H), 7.31 (ddd, 1H,
J ) 8.0, 8.0, 2.0 Hz), 7.26 (dd, 1H, J ) 7.8, 2.0 Hz), 6.71 (d,
1H, J ) 8.4 Hz), 6.64 (ddd, 1H, J ) 8.0, 8.0, 1.0 Hz). 13C NMR
(acetone-d6): δ 170.5, 167.4, 136.4, 136.0, 122.7, 119.2, 116.4,
+
59.1, 32.0, 30.6, 27.0, 23.1, 14.2, -6.9. LRMS (EI): M - CH3
) 288. HRMS (EI) M+ calcd for C21H36NO2Ga: 303.1114.
Found: 303.1117.
K2N,O-2-[(6-Meth oxycar bon ylh exylim in o)m eth yl]ph en -
ola tod im eth ylga lliu m (9g). 1H NMR (acetone-d6): δ 8.37
(s, 1H), 7.31 (ddd, 1H, J ) 8.0, 8.0, 2.0 Hz), 7.26 (dd, 1H, J )
7.6, 2.0 Hz), 6.71 (d, 1H, J ) 8.8 Hz), 6.64 (ddd, 1H, J ) 7.4,
7.4, 1.2 Hz). 13C NMR (acetone-d6): δ 173.9, 170.8, 167.6,
136.5, 136.2, 122.8, 119.2, 116.5, 58.9, 51.5, 34.1, 30.4, 26.8,
K2N ,S -1-P h e n yl-2-(2-p yr id yl)-1-e t h e n e -1-t h iola t od i-
m eth ylga lliu m (12d ): Yellow solid. 1H NMR ((CD3)2CO, 500
MHz): δ 8.46 (dd, 1H, J ) 5.47, 1.87 Hz), 8.10 (ddd, 1H, J )
8.15, 7.55, 1.76 Hz), 7.80 (m, 2H), 7.67 (ddd, 1H, J ) 8.06,
1.30, 0.83 Hz), 7.55 (ddd, 1H, J ) 7.42, 5.73, 1.33 Hz), 7.37
(m, 3H), 7.15 (s, 1H), -0.18 (s, 6H); 13C NMR (CDCl3, 125 MHz)
((CD3)2CO, 125 MHz): δ 155.6, 155.4, 146.4, 145.6, 141.5,
129.4, 128.8, 128.2, 128.1, 123.7, 121.2, -6.81. MS (EI, 70
eV): m/ z (relative intensity) 313 (M+, 2), 313 (M+, 2), 298 (73),
296 (100), 212 (86), 180 (36), 152 (12), 78 (26), 71 (17), 69 (27).
Anal. Calcd for C15H16GaNS: C, 57.73; H, 5.17; N, 4.49; S,
10.27. Found: C, 57.65; H, 5.18; N, 4.55; S, 10.04.
+
25.2, -6.9. LRMS (EI): M - CH3 ) 332. HRMS (EI) M+
calcd for C21H36NO2Ga: 332.0777. Found: 332.0778.
K2N,O-4-(P h en ylim in om et h yl)b en zen e-1,3-d iola t od i-
m eth ylga lliu m (10a ). 1H NMR (acetone-d6): δ 9.04 (s, 1H),
8.74 (s, 1H), 7.22 (t, 2H, J ) 7.6, 7.6 Hz), 7.05 (m, 1H), 7.02
(d, 2H, J ) 8.5 Hz), 6.86 (dd, 2H, J ) 8.4, 1.2 Hz), 6.74 (d, 1H,
J ) 9.0 Hz), 6.73 (d, 2H, J ) 8.5 Hz), 6.27 (d, 1H, J ) 2.4 Hz),
6.10 (dd, 1H, J ) 8.9, 2.4 Hz). 13C NMR (acetone-d6): δ 169.4,
169.0, 165.4, 137.9, 113.0, 107.0, 106.7, -6.9. LRMS (EI): M
X-r a y Str u ctu r e Deter m in a tion of 7c. Single crystals
of 7c, grown from a saturated CHCl3 solution, were mounted
on glass fibers with Paratone-N oil (Exxon) and immediately
cooled to -75 °C in a cold nitrogen gas stream on the
diffractometer. Standard peak search and indexing procedures
gave rough cell dimensions, and least-squares refinement
using 10 688 reflections yielded the cell dimensions given in
Table 2. Data were collected with an area detector by using
the measurement parameters listed in Table 2. Systematic
absences for 0kl (k + l * 2n) and hk0 (h * 2n) were consistent
with space groups Pnma and Pna21. The average values of
the structure factors suggest the centri choice Pnma, which
was verified by the successful refinement of the structure. The
measured intensities were reduced to structure factor ampli-
tudes and their esd’s by correction for background and Lorentz
and polarization effects. While corrections for crystal decay
were unnecessary, a face-indexed absorption correction was
applied, the maximum and minimum transmission factors
being 0.858 48 and 0.570 50. Systematically absent reflections
were deleted and symmetry equivalent reflections were aver-
aged to yield the set of unique data. All 10 688 data were used
in the least-squares refinement. The structure was solved
using direct methods (SHELXTL). The correct positions for
the C, O, and Ga atoms were deduced from an E-map.
Subsequent least-squares refinement and difference Fourier
calculations revealed the positions of the remaining non-
hydrogen atoms. The quantity minimized by the least-squares
program was ∑w(Fo2 - Fc2)2, where w ) [[σ(Fo2)]2 + (0.0765P)2
+
- CH3 ) 388. HRMS (FAB) M+ Calcd for C21H20NO3Ga:
404.0777. Found: 404.0777.
K2N,O-4-[(4-Meth oxyp h en yl)p h en ylim in om eth yl]ben -
zen e-1,3-d iola tod im eth ylga lliu m (10b). 1H NMR (acetone-
d6): δ 9.07 (s, 1H), 7.22 (dd, 2H, J ) 7.6, 7.6 Hz), 7.11 (d, 2H,
J ) 8.8 Hz), 7.05 (t, 1H, J ) 7.6 Hz), 6.89 (d, 2H, J ) 7.3 Hz),
6.82 (d, 2H, J ) 8.8 Hz), 6.28 (d, 1H, J ) 2.4 Hz), 6.09 (dd,
1H, J ) 8.9, 2.7 Hz), 3.75 (s, 3H), -0.48 (s, 6H). 13C NMR
(acetone-d6): δ 176.0, 170.4, 164.3, 159.9, 145.9, 137.1, 131.1,
128.7, 125.3, 124.2, 112.9, 106.8, 105.6, 54.5, -8.8. LRMS
+
(EI): M - CH3 ) 402. Anal. Calcd for C22H22NO3Ga: C,
63.19; H, 5.30; N, 3.35. Found: C, 63.13; H, 5.44; N, 3.18.
K2N,O-2-(ter t-Bu tylim in om eth yl)-4-n itr op h en ola tod i-
m eth ylga lliu m (11a ). 1H NMR (acetone-d6): δ 8.81 (s, 1H),
8.48 (d, 1H, J ) 2.9 Hz), 8.18 (dd, 1H, J ) 9.3, 3.1 Hz), 6.80
(d, 1H, J ) 9.3 Hz), 1.54 (s, 9H), -0.22 (s, 6H). 13C NMR
(acetone-d6): δ 167.4, 136.5, 133.7, 130.2, 122.1, 117.7, 61.3,
29.3, -3.7. LRMS (EI): M - CH3+ ) 305. HRMS (FAB) M -
+
CH3 Calcd for C13H19N2O3Ga: 305.0413. Found: 305.0417.
K2N,O-1-P h en yl-2-(2-p yr id yl)-1-eth en ola tod im eth ylga l-
liu m (12a ): Yellow solid. 1H NMR (CDCl3, 500 MHz): δ 7.86
(m, 3H), 7.63 (ddd, 1H, J ) 8.3, 7.3, 1.7 Hz), 7.38 (m, 3H),
7.04 (dt, 1H, 7.0, 0.98 Hz), 6.99, (ddd, 1H, 7.3, 5.7, 1.3 Hz),
5.93 (s, 1H), -0.19 (s, 6H). 13C NMR (CDCl3, 125 MHz): δ
169.2, 156.9, 143.2, 140.1, 139.0, 129.3, 128.1, 126.4, 123.3,
118.3, 93.9, -7.65. MS (EI, 70 eV): m/ z (relative intensity)
297 (M+, 5), 295 (M+, 8), 282 (68), 280 (100), 266 (8), 264 (12),
2
+ 3.6495P]-1 and P ) (Fo + 2Fc2)/3. The analytical ap-
proximations to the scattering factors were used, and all
structure factors were corrected for both real and imaginary
components of anomalous dispersion. In the final cycle of least
squares, independent anisotropic displacement factors were
refined for the non-hydrogen atoms and the aromatic, vinyl,
and methyl hydrogen atoms were fixed in “idealized” positions,
with C-H ) 0.95 Å for the aromatic and vinyl hydrogens and
C-H ) 0.98 Å for the methyl hydrogens. Successful conver-
gence was indicated by the maximum shift/error of 0.001 for
the last cycle. Final refinement parameters are given in Table
2. The largest peak in the final Fourier difference map (1.966
e Å-3) was located 0.88 Å from C10. A final analysis of
163 (8), 161 (11), 71 (90), 69 (14). Anal. Calcd for C15H16
-
GaNO: C, 60.86; H, 5.45; N, 4.73. Found: C, 60.84; H, 5.31;
N, 4.76.
K2N,O-1-(4-Hyd r oxyp h en yl)-2-(2-p yr id yl)-1-eth en ola to-
d im eth ylga lliu m (12b): Yellow solid. 1H NMR (CDCl3, 500
MHz): δ 8.89 (bs, 1H), 7.70 (dd, 1H, J ) 5.61, 1.65 Hz), 7.59
(AA′ of AA′XX′, 2H, J AX ) 8.60 Hz, J AA ) 2.48 Hz), 7.47 (ddd,
1H, J ) 8.46, 7.06, 1.85 Hz), 6.89 (dt, 1H, J ) 8.35, 1.04 Hz),
6.82 (ddd, 1H, J ) 7.09, 5.76, 1.27 Hz), 6.724 (XX′ of AA′XX′,
2H, J AX ) 8.58 Hz, J XX ) 2.47 Hz), 5.73 (s, 1H), -0.35 (s, 6H).
13C NMR (CDCl3, 125 MHz): δ 168.9, 158.5, 156.7, 142.7,

4896 Organometallics, Vol. 17, No. 22, 1998
Chesnut et al.
variance between observed and calculated structure factors
showed no apparent errors.
Hyd r olysis Stu d ies. A weighed sample of each gallium
complex was dissolved in acetone-d6 to form a 0.20 M solution.
At room temperature, H2O was added to achieve a H2O/Ga
ratio of 1000. Formation of (CH3)2GaOH was monitored by
1H NMR, with the percent hydrolysis measured by integrating
the (CH3)2Ga- peaks corresponding to the unhydrolyzed
complex and to (CH3)2GaOH. The determination of percent
hydrolysis was made 12-24 h after the addition of H2O. These
percent hydrolysis values are listed in Figure 2.
X-r a y Str u ctu r e Deter m in a tion of 12a . Single crystals
of 12a were grown and mounted the same as described for
compound 7a above. Standard peak search and indexing
procedures gave rough cell dimensions, and least-squares
refinement using 8701 reflections yielded the cell dimensions
given in Table 2. Data were collected with an area detector
by using the measurement parameters listed in Table 2.
Systematic absences for 0k0 (k * 2n) and h0l (h + l * 2n)
were consistent only with space group P21/n. The measured
intensities were reduced to structure factor amplitudes and
their esd’s by correction for background and Lorentz and
polarization effects. While corrections for crystal decay were
unnecessary, a ψ-scan absorption correction was applied, the
maximum and minimum transmission factors being 0.990 and
0.861. Systematically absent reflections were deleted and
symmetry equivalent reflections were averaged to yield the
set of unique data. All 8701 data were used in the least-
squares refinement. The structure was solved using direct
methods (SHELXTL). The correct positions for the C, N, O,
and Ga atoms were deduced from an E-map. Subsequent
least-squares refinement and difference Fourier calculations
revealed the positions of the remaining non-hydrogen atoms.
The quantity minimized by the least-squares program was
∑w(Fo2 - Fc2)2, where w ) [[σ(Fo2)]2 + (0.0138P)2 + 1.3221P]-1
Ack n ow led gm en t. We are grateful for support of
this research through grants from the National Insti-
tutes of Health (5 F32 CA71174-02), Department of the
Army (DAMD17-97-1-7292), National Cancer Institute
DHHS (PHS 5 T32 CA09067), and the Department of
Energy (DE FG02 86ER60401 and DE FG02 84ER-
60218). NMR spectra were obtained in the Varian
Oxford Instrument Center for Excellence in NMR
Laboratory. Funding for this instrumentation was pro-
vided in part from the W. M. Keck Foundation, the
National Institutes of Health (PHS 1 S10 RR104444-
01), and the National Science Foundation (NSF CHE
96-10502). Mass spectra were obtained on instruments
supported by the grants from the National Institute of
General Medical Sciences (GM 27029), the National
Institutes of Health (RR 01575), and the National
Science Foundation (PCM 8121494).
2
and P ) (Fo + 2Fc2)/3. The analytical approximations to the
scattering factors were used, and all structure factors were
corrected for both real and imaginary components of anoma-
lous dispersion. In the final cycle of least squares, independent
anisotropic displacement factors were refined for the non-
hydrogen atoms, and the aromatic, vinyl, and methyl hydrogen
atoms were fixed in “idealized” positions, with C-H ) 0.95 Å
for the aromatic and vinyl hydrogens and C-H ) 0.98 Å for
the methyl hydrogens. Successful convergence was indicated
by the maximum shift/error of 0.001 for the last cycle. Final
refinement parameters are given in Table 2. The largest peak
in the final Fourier difference map (0.330 e Å-3) was located
0.88 Å from C14. A final analysis of variance between
observed and calculated structure factors showed no apparent
errors.
Su p p or t in g In for m a t ion Ava ila b le: Full crystallo-
graphic information, atomic coordinates and isotropic displace-
ment parameters, and intramolecular bond distances and
angles can be found on pages S-2-10 for compound 7c and
pages S-11-17 for compound 12a ; experimental procedures
and characterization for all ligands can be found on pages
S-18-27 (28 pages). Ordering information can be found on
any current masthead page.
OM980413X