Within the framework of SU(3) flavor symmetry, we investigate two-body doubly charmful baryonic $$ B\to {\textbf{B}}_c{\overline{\textbf{B}}}_c^{\prime } $$
B
→
B
c
B
¯
c
′
decays, where $$ {\textbf{B}}_c{\overline{\textbf{B}}}_c^{\prime } $$
B
c
B
¯
c
′
represents the anti-triplet charmed dibaryon. We determine the SU(3)f amplitudes and calculate $$ \mathcal{B}\left({B}^{-}\to {\Xi}_c^0{\overline{\Xi}}_c^{-}\right)=\left({3.4}_{-0.9}^{+1.0}\right)\times {10}^{-5} $$
B
B
−
→
Ξ
c
0
Ξ
¯
c
−
=
3.4
−
0.9
+
1.0
×
10
−
5
and $$ \mathcal{B}\left({\overline{B}}_s^0\to {\Lambda}_c^{+}{\overline{\Xi}}_c^{-}\right)=\left({3.9}_{-1.0}^{+1.2}\right)\times {10}^{-5} $$
B
B
¯
s
0
→
Λ
c
+
Ξ
¯
c
−
=
3.9
−
1.0
+
1.2
×
10
−
5
induced by the single W-emission configuration. We find that the W-exchange amplitude, previously neglected in studies, needs to be taken into account. It can cause a destructive interfering effect with the W-emission amplitude, alleviating the significant discrepancy between the theoretical estimation and experimental data for $$ \mathcal{B}\left({\overline{B}}^0\to {\Lambda}_c^{+}{\overline{\Lambda}}_c^{-}\right) $$
B
B
¯
0
→
Λ
c
+
Λ
¯
c
−
. To test other interfering decay channels, we calculate $$ \mathcal{B}\left({\overline{B}}_s^0\to {\Xi}_c^{0\left(+\right)}{\overline{\Xi}}_c^{0\left(+\right)}\right)=\left({3.0}_{-1.1}^{+1.4}\right)\times {10}^{-4} $$
B
B
¯
s
0
→
Ξ
c
0
+
Ξ
¯
c
0
+
=
3.0
−
1.1
+
1.4
×
10
−
4
and $$ \mathcal{B}\left({\overline{B}}^0\to {\Xi}_c^0{\overline{\Xi}}_c^0\right)=\left({1.5}_{-0.6}^{+0.7}\right)\times {10}^{-5} $$
B
B
¯
0
→
Ξ
c
0
Ξ
¯
c
0
=
1.5
−
0.6
+
0.7
×
10
−
5
. We estimate non-zero branching fractions for the pure W-exchange decay channels, specifically $$ \mathcal{B}\left({\overline{B}}_s^0\to {\Lambda}_c^{+}{\overline{\Lambda}}_c^{-}\right)=\left({8.1}_{-1.5}^{+1.7}\right)\times {10}^{-5} $$
B
B
¯
s
0
→
Λ
c
+
Λ
¯
c
−
=
8.1
−
1.5
+
1.7
×
10
−
5
and $$ \mathcal{B}\left({\overline{B}}^0\to {\Xi}_c^{+}{\overline{\Xi}}_c^{-}\right)=\left(3.0\pm 0.6\right)\times {10}^{-6} $$
B
B
¯
0
→
Ξ
c
+
Ξ
¯
c
−
=
3.0
±
0.6
×
10
−
6
. Additionally, we predict $$ \mathcal{B}\left({B}_c^{+}\to {\Xi}_c^{+}{\overline{\Xi}}_c^0\right)=\left({2.8}_{-0.7}^{+0.9}\right)\times {10}^{-4} $$
B
B
c
+
→
Ξ
c
+
Ξ
¯
c
0
=
2.8
−
0.7
+
0.9
×
10
−
4
and $$ \mathcal{B}\left({B}_c^{+}\to {\Lambda}_c^{+}{\overline{\Xi}}_c^0\right)=\left({1.6}_{-0.4}^{+0.5}\right)\times {10}^{-5} $$
B
B
c
+
→
Λ
c
+
Ξ
¯
c
0
=
1.6
−
0.4
+
0.5
×
10
−
5
, which are accessible to experimental facilities such as LHCb.