diff --git a/tools/vasp/reactions/find_molecular.py b/tools/vasp/reactions/find-molecules-from-table.py
similarity index 100%
rename from tools/vasp/reactions/find_molecular.py
rename to tools/vasp/reactions/find-molecules-from-table.py
diff --git a/tools/vasp/reactions/find-molecules.py b/tools/vasp/reactions/find-molecules.py
new file mode 100644
index 0000000..da9b528
--- /dev/null
+++ b/tools/vasp/reactions/find-molecules.py
@@ -0,0 +1,80 @@
+"""
+@todo: assign atom types
+"""
+import os, sys
+import numpy as np
+import ase
+import ase.io
+from ase.neighborlist import NeighborList
+from ase.neighborlist import neighbor_list
+from ase.data import covalent_radii, atomic_numbers
+
+def reax_bond_order(a, b, r, ffield):
+    bo = math.exp(pbo1*math.power(r/r0, pbo2))
+    return bo
+
+def build_cutoff_table(atps):
+    scale = 1.2
+    cutoff_table = {}
+    for i in range(len(atps)):
+        for j in range(i, len(atps)):
+            ri = covalent_radii[atomic_numbers[atps[i]]]
+            rj = covalent_radii[atomic_numbers[atps[j]]]
+            r0 = scale*(ri+rj)
+            cutoff_table[(atps[i], atps[j])] = r0
+    return cutoff_table
+
+def get_lists(atoms, cutoff_table):
+
+    # build neighbor list
+    nl = []
+    for i in range(len(atoms)):
+        nl.append([])
+
+    tokens_i, tokens_j = neighbor_list('ij', atoms, cutoff_table)
+    for i in range(len(tokens_i)):
+        nl[tokens_i[i]].append(tokens_j[i])
+
+    return nl
+
+def get_molecule_id(atom_id, mol_id, atom_visits, molecules):
+    if atom_visits[atom_id][2] < 0:
+        atom_visits[atom_id][2] = mol_id
+        molecules[mol_id].append(atom_id)
+        for partner in atom_visits[atom_id][1]:
+            get_molecule_id(partner, mol_id, atom_visits, molecules)
+
+def find_molecules(atoms, nl):
+    molecules = []
+    for i in range(len(atoms)):
+        molecules.append([])
+
+    atom_visits = []
+    for i in range(len(atoms)):
+        atom_visits.append([atoms[i].symbol, nl[i], -1])
+
+    for i in range(len(atoms)):
+        get_molecule_id(i,i, atom_visits, molecules)
+
+    molecules = [i for i in molecules if i]
+    return molecules
+
+def output_molecules(molecules, atoms):
+    o = open('mol.dat', 'w')
+    for i in molecules:
+        for j in i:
+            o.write('%5d '%j)
+        o.write('\n')
+    o.close()
+
+if len(sys.argv) > 1:
+    fname = sys.argv[1]
+    atoms = ase.io.read(fname)
+    atps = list(set(atoms.get_chemical_symbols()))
+    cutoff_table = build_cutoff_table(atps)
+    #cutoff_table[('O', 'Li')] = 0.0
+    #cutoff_table[('Li', 'O')] = 0.0
+    nl = get_lists(atoms, cutoff_table)
+    molecules = find_molecules(atoms, nl)
+    output_molecules(molecules, atoms)
+