PHYSICAL REVIEW C 100 , 014311 ( 2019 ) Lifetimes and shape-coexisting states of 99 Zr

Spagnoletti, P.; Simpson, G.; Kisyov, S.; Bucurescu, D.; Regis, J-M; Saed-Samii, N.; Blanc, A.; Jentschel, M.; Koster, U.; Mutti, P.; Soldner, T.; de France, G.; Ur, C. A.; Urban, W.; Bruce, A. M.; Bernards, C.; Drouet, F.; Fraile, L. M.; Gaffney, L. P.; Ghita, D. G.; Ilieva, S.; Jolie, J.; Korten, W.; Kroll, T.; Lalkovski, S.; Larijarni, C.; Lica, R.; Mach, H.; Marginean, N.; Paziy, V; Podolyak, Zs; Regan, P. H.; Scheck, M.; Smith, J. F.; Thiamova, G.; Townsley, C.; Vancraeyenest, A.; Vedia, V.; Warr, N.; Werner, V; Zielinska, M.


I. INTRODUCTION
Nuclei of the neutron-rich A ≈ 100 region possess a wide variety of shape phenomena. A rapid change in ground-state deformation between the spherical 96 Zr, weakly deformed 98 Zr [1], and strongly deformed 100 Zr [2] is well known. This discovery was surprising as the properties of the atomic nucleus generally evolve smoothly as a function of neutron or proton number. The Z = 38 (π p 3/2 ) and 40 (π p 1/2 ) and the N = 50 (νg 9/2 ), 56 (νd 5/2 ), and 58 (νs 1/2 ) spherical subshell closures, combined with a weak integrated residual proton-neutron interaction, mean that from N = 50 to 58 the low energy structure of the Sr and Zr isotopes resembles that of a semimagic nucleus. At N = 60 a strongly (prolate) deformed ground-state rotational band develops and forms the first few states. This shape change is more gradual across N = 58-60 for the higher-Z Mo-Pd nuclei of the region, which exhibit triaxiality [3][4][5][6]. At lower Z values the N 60 nuclei 96,98,100 Kr have recently been shown to possess considerably less collectivity, due to reduced π g 9/2 occupation and prolate-oblate shape coexistence [7,8]. The rapid nature of this ground-state deformation change has made this region a challenge to study theoretically. Furthermore, these neutron-rich nuclei are often far from stability and difficult to access experimentally.
A variety of theoretical explanations exist to explain this shape change including the crossing of two different meanfield configurations; one governed by spherical shell effects and the other a deformed intruder containing multiple particlehole excitations, dominated by the integrated residual protonneutron interaction [9,10]. Experimental evidence for coexisting spherical and deformed configurations exists for 96-98 Sr [11,12], 96-104 Y [13][14][15], and 98-100 Zr [10,[16][17][18][19]. A strong interaction between occupied π g 9/2 -νg 7/2 spin-orbit partners has been proposed to play a major role in obliterating the N = 56, 58 and Z = 38, 40 spherical subshell closures, leading to collective motion [20][21][22]. Occupation of highj, low-deformation driving νh 11/2 orbitals [21], along with the presence of deformed shell gaps at Z = 38, N = 60 [4,13], has also been shown to help stabilize deformation here. The rapid onset of the ground-state deformation is an example of quantum shape phase transition [23]. Recently type-II shell evolution [24] has been proposed to explain the onset of collectivity in this region [25]. In this mechanism specific particle-hole excitations for excited levels modify the underlying shellstructure due to the changed occupation probabilities altering the strength of the tensor force.
For the Zr isotopes the latter mechanism occurs once the νd 5/2 subshell approaches full occupation and neutrons can then first occupy the νg 7/2 and νh 11/2 orbits. Proton particlehole excitations provide initial π g 9/2 subshell occupation. These then interact strongly and attractively with neutrons situated in both the νg 7/2 spin-orbit partner and νh 11/2 orbits. The tensor and central forces act coherently on the νg 7/2 orbit, lowering its monopole energy, which in turn results in higher occupancy. Although the tensor and central forces act incoherently for the π g 9/2 -νh 11/2 coupling, the net interaction is still attractive. The energy of the π g 9/2 -π p 1/2 subshell gap is then reduced, further increasing its filling. Higher occupancy of the π g 9/2 orbit leads to the creation of more holes in the π p f shell, further increasing collectivity. Evidence for this feedback mechanism presents itself beyond N = 56, once the νd 5/2 subshell approaches full occupation and neutrons can first enter the νg 7/2 and νh 11/2 orbits. Quasirotational bands appear at 1581.6 and 854.0 keV in 96,98 Zr, which drop further in energy and increase in collectivity in 100 Zr, to become the ground-state rotational band.
It is of particular interest to study the properties of shapecoexisting structures at the spherical-deformed border (N = 59), which may allow additional insights into the underlying mechanisms driving the shape change. To date no coherent theoretical explanation exists describing all shape coexisting features found in this region, even though it contains some of the best known cases in the whole nuclear landscape [26]. In 99 Zr low-lying spherical states coexist with well-deformed rigid rotational bands situated at 0.6-1.0 MeV [10,17,18,27]. An absence of reliable and pertinent experimental data on the rotational bands of 99 Zr has so far hampered any theoretical interpretations. Different quadrupole moments have been reported for the ν9/2 [404] and ν3/2[541] bands of 99 Zr [17,18], implying that the rigidity of its core is sensitive to the occupation of specific orbitals. Furthermore, there is, to date, no evidence for oblate shapes expected to exist and play a significant role in the rapid onset of ground-state deformation here. For this reason we have chosen to study the shapes of the rotational bands in 99 Zr.
Measurements of the electromagnetic transition rates between different states in the bands of odd-A nuclei can give crucial information on collectivity, or help identify the unpaired orbital. The reduced E 2 transition rate, B(E 2), is directly related to quadrupole collectivity. This quantity can be derived from the measured lifetime of a stretched intraband E 2 transition. Alternatively, the underlying single-particle structure of a band can be probed by measurements of intraband B(M1) values, which are often orbital dependent. In the present study the lifetimes of excited states in 99 Zr have been measured. These nuclei were populated following the cold-neutron induced fission of 235 U. Previously, the β decay of mass-separated 99 Y ions was used to populate 99 Zr for lifetime measurements [10]. The 5/2 + spin of the ground state of 99 Y [15], originating from the occupation of a π 5/2[422] Nilsson orbital, permitted only limited access to members of the rotational bands of 99 Zr. The prompt-fission reaction produces nuclei, on average, with intermediate-spin states (6h-8h), allowing lifetime measurements of levels with higher spins than those fed following β decay.

II. EXPERIMENT
Lifetimes of excited states in 99 Zr were measured via the fast-timing technique. This uses γ (E , t )-γ (E , t ) coincidences for direct timing measurements and relies on the good time resolution (≈200 ps) of LaBr 3 (Ce) detectors to access the lifetimes of states in the 10 ps-to-ns time range. Cold-neutron induced fission of 235 U produces secondary neutron-rich nuclei in the A ≈ 100 region with average spins of 6h-8h. It is therefore a suitable reaction to populate the excited states of interest in 99 Zr, especially as this nucleus is situated at one peak of the fission-yield distribution. As some 100 fission fragments are reasonably well produced in this reaction, the modest energy resolution of the LaBr 3 (Ce) detectors is not sufficient to uniquely select a γ -ray cascade of interest. This requires additional γ -ray detection using Ge detectors with their superior energy resolution.
The experiment was performed at the PF1B cold-neutron guide [28] of the high-flux reactor of the Institut Laue-Langevin, Grenoble, within the framework of the EXILL-FATIMA campaign [29]. The collimated neutron beam, with a flux ≈ 10 8 n/(cm 2 s), was ≈1 cm 2 in area [30] and induced fission in a 0.8-mg 235 UO 2 (0.675 mg 235 U) target. The target was sandwiched between two 25-μm-thick Be backings, in order to stop the fission fragments in a time of just a few ps. This minimized the Doppler broadening of any emitted prompt γ rays. The target was placed at the center of the EXILL-FATIMA γ -ray detector array, which consisted of 8 EXOGAM clover Ge and 16 LaBr 3 (Ce) detectors [29]. The data were acquired during a two-week measurement period and sorted offline into γ (Ge)-γ (LaBr 3 )-γ (LaBr 3 ) coincidence events, occurring within a 120-ns time window. Timing signals of the LaBr 3 (Ce) detectors were recorded in analog FIG. 1. Partial decay scheme of 99 Zr. Transition and level energies are taken from Ref. [18]. Some known weak decay branches are not shown. time-to-amplitude converters (TACs) with a 50 ns range. More details on the experiment can be found in Ref. [29].

III. RESULTS
A partial level scheme of 99 Zr, containing the levels and decays of interest for the present work, is shown in Fig. 1. Here the level and transition energies are taken from Ref. [31].
The lifetime of the 121.7-keV level has been reported to be measured several times [31] and, therefore, constitutes a good test case. Furthermore several lifetimes measured from this data set have been reported previously [32][33][34]. The 192.6-536.0-121.7-keV cascade was used for this. The spectra presented in Fig. 2 were obtained by gating on the 192.6-keV transition (blue), using a background gate at 208 keV (red), with the Ge detectors, and by gating on the 536.0-keV transition with the LaBr 3 (Ce) detectors. The corresponding time spectra (see Fig. 3) were produced with an additional gate using the LaBr(Ce) detectors on the 121-keV transition and shows that the background is not entirely prompt and therefore contributes to the slow component of the time spectrum. The background is subtracted and the slope is measured. This was performed for both the delayed and antidelayed time spectra and a weighted average is given. The delayed time spectrum corresponds to a time-to-amplitude converter (TAC) being started with a transition which feeds the level of interest and stopped by a transition decaying from the level of interest. The  I. Gates set on Ge and LaBr 3 (Ce) detectors for lifetime (τ ) measurements and the results obtained. The subscripts i j and ji correspond to the gate orders start/stop and stop/start, respectively. Lifetimes above 1 ns were determined using the slope (SL) method and lifetimes below 1 ns were determined using the centroid difference (CD) method. without employing any background subtraction. This method was also employed to measure the lifetime of the 2 + 1 state in 146 Ba. A value of τ = 1260(40) ps is obtained, which agrees well with the value of τ = 1240(40) ps reported in the compilation of Ref. [35] Measurements of the lifetimes of four other states of 99 Zr were possible. The various gates used are listed in Table I, along with the lifetimes obtained. For the 850.5-keV state a quadruple Ge-Ge-LaBr 3 -LaBr 3 coincidence was required to achieve the necessary selectivity. A summary of the results obtained is presented in Table I difference by the equation where FEP stands for full-energy peak and PRD is the prompt response difference. The PRD describes the combined γ -γ time walk of the setup [36].
In the case where the background contribution to the time spectra are not negligible, the centroid difference between the time distributions corresponding to the full-energy peaks, C FEP , is related the the experimental centroid difference, C exp , by the relation where The term C bg is the centroid difference of the background and is determined analytically for both transitions. The analytical corrections for both the feeding and decaying transitions are shown in Figs. 6 and 7, respectively. The term P/B is the peak-to-background ratio. A thorough description of this method is presented in Refs. [33,34] Transition branching ratios from [31] and corrections for internal conversion [37] were used to convert the measured level lifetimes into partial stretched E 2 lifetimes (τ E 2 ). Reduced transition rates were then calculated using the formula From the B(E 2) values transition quadrupole moments, Q t , can be derived in the rigid-rotor framework using the formula where J i is the initial spin, J f is the final spin, and K is the projection of the aligned angular momentum on the symmetry axis. Deformation parameters, β 2 , can be determined using with Q t = Q 0 . For J = 1 transitions the experimental partial M1 lifetime (τ M1 ) is related to the reduced transition rate by In the case of the (3/2 + ) band built on either the 575.7- [18] or 657.9-keV [27] states, conflicting spin assignments exist. The present work adopts the J π assignments from Ref. [18] and provides experimental evidence to support this choice in Sec. IV.   Table III.

5-keV levels
The exact nature of the 575.7-and 657.9-keV levels is uncertain. A positive parity was assigned to these states based on measured log f t values from the β decay of 99 Y [10,40]. In [17] these two levels were presented as the bandheads of two separate K = 3/2 + and 5/2 + decoupled bands. This was then revised to them being the (3/2 + ) and (5/2 + ) members of the same K = 3/2 + band [18]. This interpretation is supported by an observed transition between the K = 9/2 + isomer at 1038.5 keV and the 657.9-keV level with energy 381 keV. However, Lhersonneau et al. later noted that the 82.2-keV K + 1 → K transition connecting them appeared to be too weak to be an intraband transition [40]. In a separate work the 657.9-keV state was proposed to be a 3/2 + bandhead, with mixed ν3/2 [ [27]. The mixed nature of this band was proposed due to the poor agreement between B(M1)/B(E 2) values derived from J=1 J=2 branching ratios and theoretically predicted ones [27].

575.7-keV level was not reported
The 381-keV transition was rejected in the evaluation by [31] as it is inconsistent with the adopted J π = 3/2 + assigned to the 657.9-keV level. Figure 8 shows a coincident γ -ray spectrum. This figure corresponds to a gate on the 218-keV transition, which feeds the K = 9/2 + isomer, and gates on the 536.0-and 121.7-keV transitions, which correspond to the cascade of decays from the 657.9-keV level. The 381-keV transition is clearly present along with the 251-keV transition that feeds the J π = 11/2 + above the K = 9/2 + isomer. Therefore a J π = 3/2 + assignment for the 657.9-keV level is rejected and the J π = 5/2 + assignment as suggested in Ref. [18] is proposed to be correct. The disparity in intensity between the 192.6-and 188.5-keV transitions is due to the 215.5-keV transition which, populating the 850.5-keV level, has an energy similar to the 218-keV coincidence gate.
The reduced M1 transition rates obtained in the present work, B(M1) = 0.032(10) and 0.042(21) W.u. for the 192.6and 82.2-keV transitions, respectively, are the same, within errors. This implies the 850.5-, 657.9-, and 575.7-keV levels have a similar structure, or are all members of the same band, in agreement with [18]. The low intensity of the 82.2-keV intraband transition is therefore a consequence of other strong decay paths out of the 657.9-keV level. Its energy is likely perturbed by interactions with other nearby (3/2 + ) states, or the 3/2 + 1 level. The band containing the 575.7-, 657.9-, and 850.5-keV levels has been proposed to have a ν3/2[411] configuration, which has a significant νg 7/2 component [18]. The log f t values derived for β feeding to these levels are 6.3, 6.7, and 6.6, respectively [40]. These values are much higher than those of an allowed Gamow-Teller transition, likely due to the strong spherical νd 5/2 single-particle component and quasihole nature of the ν3/2 [411] orbital. The collectivity present in the band will also fragment the νg 7/2 strength across several levels, resulting in higher log f t values. The shape coexistence present in 99 Zr and mixed configurations means that it is unclear if asymptotic Nilsson orbital labels are applicable. Nevertheless, a ν3/2 [411] label is kept in the article for consistency with the existing literature.
The values of β 2 = 0.34(1) and 0.26(3) derived for the (11/2 − ) and (15/2 − ) members of the same ν3/2[541] band of 99 Zr in the present work are inconsistent. The latter β 2 value, however, agrees with β 2 = 0.28(1) measured for this band's (19/2 − ) and (23/2 − ) members [17]. It is also close to β 2 = 0.32(2) determined for the same band in the isotone 97 Sr [17]. The change in quadrupole moment across the 821.6-keV state, with spin (11/2 − ), may indicate a crossing with a Nilsson orbital of the same spin. A deformation of β 2 = 0.356 (8) has been reported for the (2 + ) state of 100 Zr [41], showing that the collectivities of members of the ν3/2 [541] band are close to that of the deformed even-even ground-state core. Simplistically this band can be thought of as a ν3/2[541] hole coupled to a 100 Zr core, with the hole acting to reduce the quadrupole moment.  99 Zr behave differently, and it seems that the occupation of specific orbitals in the underlying core provokes a structural change in the core and affects the rotational properties. The E (4 + 1 )/E (2 + 1 ) ratio of 100 Zr is 2.66, characteristic of a transitional or γ -unstable nucleus, whereas those of 98 Sr and 102 Zr are 3.00 and 3.15, respectively, close to the rigid-rotor limit.

C. Moments of inertia
The soft nature of the ν3/2[411] band in 99 Zr provides evidence that occupation of the νg 7/2 and/or νd 5/2 orbitals does not reduce the pairing correlations present in the core as much as in the bands of 101,103 Zr. The stiffer nature of the core and larger deformation of the ν3/2[541] band points towards occupation of the νh 11/2 orbital in reducing the pairing correlations present in the core and a move towards a rigid, well-deformed rotor regime. Differences in the kinematic moment of inertia of the ν3/2[451] band in 99 Zr are presented in Fig. 10. The band has a near constant moment of inertia with the exception of the low-spin levels, which coincide with the large increase in B(E 2) values. This indicates that rigid rotation is is not a suitable description for the low-spin levels within the ν3/2[451] band.

D. IBFM-1
As a result of the recent measurement [42] of the magnetic moment, μ, of the 7/2 + 1 state of 99 Zr, the three lowest-lying states of this nucleus now have a rather complete experimental characterization, with experimentally determined magnetic moments and electromagnetic transition rates [17,18,27]. Since the lowest-lying states in an odd-mass nucleus are mainly of one-quasiparticle nature, it was decided to investigate their description with a simple model, the interacting boson-fermion model (IBFM) in its version IBFM-1, which does not distinguish between protons and neutrons [43,44]. In this model, the odd nucleus is described by coupling an odd nucleon, allowed to occupy single-particle shell-model orbitals, to an even-even core described by the interaction boson model 1 (IBM-1) [45,46]. The description of the known spectroscopic properties of the lowest-lying states allowed a fine tuning of the IBFM-1 parameters [42] (see also Table IV).
Here we use the results of these calculations to investigate their ability to describe the properties of other excited states, including those determined in this experiment. It should, however, be emphasized that the IBFM-1 description has strong limitations. Nuclei in this region of rapid shape transition, around N = 60, have been found to exhibit shape coexistence phenomena, which cannot be accounted for by this model. The fast structure evolution around N = 60 also makes it difficult to choose an appropriate even-even core nucleus for 99 Zr as the ground-state deformation changes drastically from 98 Zr to 100 Zr, its two even-even neighbors. Here, 100 Zr was chosen as a core due to the similarity between the energies of its ground-state band members and the energy distribution of the low-lying states in 99 Zr. The IBM-1 description employed for the core was taken from Ref [47]. The odd neutron coupled to the core was allowed to span the single-particle orbits of the N = 50-82 valence space (s 1/2 , d 3/2 , d 5/2 , g 7/2 , and h 11/2 ). The IBFM-1 calculations are similar to those performed in Ref. [48] for the isotonic nucleus 97 Sr, which presents structure similarities with 99 Zr. The main difference from the calculations in Ref. [48] is the inclusion of the νh 11/2 orbital in the present calculations, and the use of the same IBFM-1 parameters for the description of both positive and negative parity states.
The single-particle energies for nuclei in this region are not known; they were adjusted to reasonably reproduce the low-energy part of the experimental level scheme. The values adopted for these calculations are given in Table IV, together with quasiparticle energies obtained from BCS calculations with a pairing parameter = 1.5 MeV. The values chosen for the boson-fermion interaction parameters [43,44] were A 0 = 0.08 MeV, 0 = 0.3 MeV, 0 = 2.9 MeV 2 , andhω = 1.5 MeV.
Electromagnetic E 2 transition rates were calculated using a boson charge e B = 0.159 eb which describes the B(E 2) values in 100 Zr [47] and a fermion charge e F with the same value. For the M1 transitions we used the simplest transfer operator, with gyromagnetic ratios g d = 0.4μ N for the d-boson g factor, and g s = −2.68μ N for the fermion, which is 70% of the free neutron value to account for core polarization effects, and a value g l = 0 (the odd nucleon is a neutron).  highlights some differences between the ENSDF adopted level scheme [31] and the spin assignments of Ref. [18] (see also Fig. 1), showing two distinct bandlike structures: one based on the 576-keV 3/2 + state, and the remaining part of the (A, a) structure above the 1323-keV 9/2 + state.
In Fig. 11 correspondences between calculated and experimental levels are indicated. These are based not only on the excitation energy, but also on the description of the experimental γ -decay patterns of the states (branching ratios and, when known, reduced transition probabilities). The correspondence between calculated and experimental levels is indicated by the order number of the calculated states of each spin. Tentative associations are indicated by placing this order number within parentheses.
For positive parities, besides the lowest energy states, a one-to-one correspondence with calculated states could be established in this way for a number of states with spin 3/2 and 5/2 in the energy range ≈600-1000 keV Also, for the higherspin members of the (A, a) structure, theoretical counterparts have been proposed. Some difficulties were encountered in the description of the decay patterns of the states at 850, 1066, and 1323 keV, therefore their theoretical assignment is only tentative. Table V shows the comparison between calculated and experimental transition rates. While for the lower-energy states the agreement is good, for the 850-keV 7/2 + state it is worse, especially its E 2 branch towards the 576-keV 3/2 + state is predicted less collective, with a B(E 2) value 3-4 times smaller than the experimental value. The decay pattern of the 1323-keV 9/2 + state towards the 850-keV 7/2 + and 1066-keV 9/2 + states [31] also could not be well reproduced (therefore, these states are only tentatively associated with calculated ones). It appears that the connection between the two level structures [band (A, a) and the band based on the 576-keV 3/2 + state[ cannot be well described by this simple model (in terms of these calculations, one would need a 014311-8  Fig. 11. All J = 0, 1 transitions were considered to be pure M1 in the experimental analysis. different mixing of the d 5/2 and g 7/2 orbitals in this energy region). The experimental band labeled "Band C" in Fig. 11, which was firmly assigned as ν9/2[404] band originating from the g 9/2 orbital [31], cannot be described with the present approach.
For the negative parity states, only the fully aligned sequence of states 11/2 − , 15/2 − , and 19/2 − are reasonably described, while the states of lower spins are predicted to be higher in energy. This may be due to a number of factors, such as the considerable change of structure around N = 60, or a drastic change in the monopole energy of the νh 11/2 orbital.

E. Appearance of collectivity in 99 Zr
The reasonable reproduction of the level scheme of 99 Zr by the IBFM-1 calculations allows some conclusions concerning the onset of collective behavior in this nucleus. Collective modes occur when the integrated proton-neutron interaction overcomes the spherical symmetry restoring pairing force. Quantitatively this happens when − < λ, where λ is the energy of the Fermi surface. The BCS calculation in Sec. IV D gives λ = 0.327 MeV using N = 1.5 MeV. Examining the single-particle energies in Table IV shows that the νs 1/2 , νd 3/2 , and νd 5/2 orbits fulfill the condition for a collective regime whereas − ≈ λ for the νg 7/2 orbit. These four orbits form a pseudo-SU(3) algebraic group, corresponding to a deformed rotor. One notes that the L = J = 2 s 1/2 -d 5/2 and d 3/2 -g 7/2 couplings are quasi-SU(3) blocks.
From Table IV one can see that the four positive-parity orbitals chosen in this work become almost degenerate in 99 Zr, after following a relatively smooth evolution with the number of neutrons in the isotopes from 91 Zr to 97 Zr (experimental data from [49]). This behavior is similar to that found for the isotonic nucleus 97 Sr [10]. The near-degenerate nature of the four neutron quasiparticle energies listed would not be expected to generate the strong collectivity present towards the bottom of the ν3/2 [541] and ν3/2 [411] bands. Multiquasiparticle bands are expected at an energy of ∼2 N , much higher than the bandhead energies. Another mechanism must be present which significantly changes the nuclear structure, most likely type-II shell evolution. In particular, the poor agreement between the calculated and experimental members of the ν3/2[541] band may be due to the monopole evolution of the νh 11/2 orbit, known to be shifted in energy by type-II shell evolution [25].
As observed above, with N increasing from 51 ( 91 Zr) to 57 ( 97 Zr), the single-particle energies of the positive-parity orbitals evolve rather smoothly as the νd 5/2 and νs 1/2 orbits are filled, while the energy of the νh 11/2 orbit increases. The νh 11/2 orbit rises in energy with increasing νd 5/2 occupation due to the repulsive nature of the tensor force for two alike l + s orbits. At N = 59 a drastic change in the single-particle energies occurs, with the overall separation of all orbits greatly reduced. This coincides with significant occupation of π g 9/2 , νg 7/2 , and νh 11/2 orbits, providing evidence that monopole evolution occurs in collective regions.

V. CONCLUSION
Lifetimes of states in the ν3/2[541] and ν3/2[411] rotational bands of the nucleus 99 Zr have been measured using the fast-timing technique. The 99 Zr nuclei were populated following the neutron-induced fission of a 235 U target during the EXILL-FATIMA campaign. The ν3/2 [541] band was deduced to possess quadrupole deformations of β 2 = 0.34(1) and 0.26(3) at spins of (11/2 − ) and (15/2 − ), whereas the (7/2 + ) member of the 3/2[411] band has β 2 = 0.32(3). IBFM-1 calculations reasonably reproduce characteristics of many of the known levels of 99 Zr. An exception are the energies of the "antialigned" levels (of spin lower that 11/2) of the negative parity band, which in these calculations has a pure νh 11/2 origin. Type-II shell evolution is proposed to play a major role in the creation of low-lying rotational bands of 99 Zr.