Novel phenoxyacetic herbicides synthesized from longifolene-derived primary amine for sustainable weed management

Abstract

Because of low water solubility, herbicides containing a phenoxy acid group, such as 2,4-dichlorophenoxyacetic acid (2,4-D) and 2-methyl-4-chlorophenoxyacetic acid (MCPA), are applied with an amine, like dimethylamine (DMA) and isopropylamine (IPA), to form ammonium salts. However, the use of amine poses substantial health and environmental risks during manufacturing and utilization. The development of non-toxic high-performance herbicidal formulations using natural compounds is therefore highly desired but remains limited. In this work, three longifolene-derived ammonium phenoxyacetates and one glyphosate were synthesized and characterized. Their herbicidal activities were evaluated against Lolium multiflorum Lam. and Brassica campestris. The results showed that almost all target compounds exhibited higher herbicidal activity than DMA or IPA formulations prepared by their corresponding commercial herbicides. Particularly, compounds 6b and 6c containing the Cl atom tended to be the most active candidates, especially with notable half maximal inhibitory concentrations (IC₅₀) values of around 0.0002 mmol L⁻¹ against the root and shoot growth of Brassica campestris, which both showed complete inhibition for Lolium multiflorum Lam. root growth and Brassica campestris shoot growth at concentrations of 0.039 and 0.156 mmol L⁻¹, respectively. In addition, compound 6c showed a good broad-spectrum herbicidal effect on the root growth of 6 different weeds, especially on rice, with an IC₅₀ of 0.000085 mmol L⁻¹. It is suggested that compounds 6b and 6c could be considered as promising botanical herbicides for sustainable weed management.

---

Introduction

Herbicides, known as weed killers, have long provided effective solutions in agricultural practices.¹ Despite their benefits for world food production, there are several downsides. Concerns about environmental and health issues have amplified because of the presence of supplemental chemicals, which are added to herbicide formulations to modify the herbicide's properties.² Particularly, amines, including dimethylamine (DMA) and isopropylamine (IPA), are frequently applied in salt formulations of well-known phenoxyacetic acid herbicides [i.e., phenoxyacetic acid (PA), 2,4-dichlorophenoxyacetic acid (2,4-D) and 2-methyl-4-chlorophenoxyacetic acid (MCPA)] to increase herbicide solubility.^2,3 However, amines tend to have higher vapor pressures than the corresponding herbicides, could potentially volatilize from herbicide-amine salts and enter atmosphere, where they are a hazard to the social environment and living organism.^4–6 Besides, volatilized amines may additionally promote the loss of active ingredient of herbicide, which would result in heighted off-target drift damage and other serious problems.^7–10 Given the defects of the currently used commercial preparations, it is urgent to develop novel efficient, eco-friendly and low-toxicity herbicide substitutes for sustainable weed management.

As an alternative, botanical herbicides, mainly extracted or derived from plants, are gaining attention. Natural products, especially plant metabolites, have been favourable in botanical herbicide preparation owing to their unique chemical structures and diverse biological properties.^11,12 Longifolene, a naturally occurring tricyclic sesquiterpene, is the primary component of heavy turpentine.¹³ As the byproduct in the production of rosin and turpentine from pine oleoresin, the sustainable biomass resource longifolene has the advantage of good bioactivities,^14,15 is reported to be used as a versatile raw material for extensive applications in many fields,^15–23 but there are few studies on the exploration of longifolene derivatives for agricultural purposes. Notably, in our previous study, ω-aminomethyl longifolene (compound 5) with a primary amine group and a series of longifolene-derived primary amine carboxylates were synthesized via derivation of volatile longifolene, and some of them displayed significant herbicidal activity against Lolium multiflorum Lam. and Brassica campestris even at low doses.^22,23 It is worth mentioning that the half maximal inhibitory concentration (IC₅₀) values of the most active compound against the root and shoot growth of Lolium multiflorum Lam. and Brassica campestris were around 0.010 and 0.023 mmol L⁻¹.²³

In a continuous study on the high-value-added exploration of longifolene in sustainable agriculture, three longifolene-derived phenoxyacetates and one glyphosate were synthesized from compound 5 and four globally commercial herbicides, namely PA, 2,4-D, MCPA and glyphosate. The use of compound 5 instead of hazardous amines is expected to reduce the harm caused by amine volatilization, and improve the herbicidal performance compared to commercially used herbicidal formulations or to achieve the same effects at a lower dosage. In addition to the synthesis and structural analysis, the herbicidal activity of target compounds against Lolium multiflorum Lam. and Brassica campestris were evaluated. Besides, the herbicidal spectrum experiment was also tested and described. Overall, this work is likely to facilitate the development of novel high-efficacy natural-based herbicides and expand the utilization of longifolene-derived compounds in agricultural fields.

Results and discussion

Synthesis and characterization

As shown in Scheme 1, target compounds 6a–6d can be prepared via a five-step synthetic route using longifolene (compound 1) as the starting material. Intermediates 2 and 3 were synthesized according to the methods that previously reported.^24,25 Compound 3 with yield of 94.2% was obtained through Prins and halogenation reaction from compound 1.²⁵ In the presence of potassium phthalimide, compound 3 reacted with dimethyl formamide (DMF) at 110 °C for 2 h, and compound 4 was produced with yield of 82.0%.^22,23 Subsequently, compound 5, yielding 94.1%, was synthesized by modification of the route based on Gabriel synthesis.²⁶ Finally, compounds 6a–6d were prepared from the reaction between compound 5 and the corresponding phenoxyacetic acids (PA, 2,4-D and MCPA) and glyphosate, respectively. These compounds were confirmed by FTIR, ¹H and ¹³C NMR spectroscopy, and HRMS. In the FTIR spectra, the peak at 3100–3000 cm⁻¹ was characteristic of the C–H stretching vibration band in the aromatic ring. The peaks in the ranges of 2958–2866 cm⁻¹, 1685–1635 cm⁻¹, 1593–1490 cm⁻¹ and 1390–1300 cm⁻¹ represented the stretching vibration band of C–H alkane groups, the stretching vibration band of the C=O bonds of the carboxyl group, the stretching vibration band of the C=C bonds of an aromatic ring, and the stretching vibration band of C–N bonds, respectively (Fig. S1–S4, SI). In the ¹H NMR spectra, three singlets at δ 0.92–0.83 ppm revealed the presence of three isolated methyl groups in ω-aminomethyl longifolene moiety. The triple at δ 5.04–4.86 ppm was assigned to the signal of H-13 and the singlet at δ 4.67–4.39 ppm belonged to the hydrogen proton of the isolated methene hydrogen proton in phenoxyacetate moiety. The chemical shifts at 7.35–6.81 ppm were attributed to hydrogen protons on the aromatic ring (Fig. S5–S8). In the ¹³C NMR spectra, peaks ranging from δ 165.32–106.97 ppm were assigned to the phenyl and carbon–carbon double bond. The peak with δ 175.14–173.94 ppm was attributed to carbonyl, and peaks with δ 68.66–16.34 ppm were assigned to the p-menthane (Fig. S9–S12). The total number of hydrogen and carbon atoms is consistent with that of the corresponding compounds, and it was further confirmed by HRMS that compounds 6a–6d are longifolene-derived ammonium phenoxyacetates and glyphosate (Fig. S13–S16).

Scheme 1 Synthetic route of target compounds 6a–6d (i) CH₃COOH, (HCHO)_n, reflux, 24 h; (ii) CH₃COCl, reflux, 1 h; (iii) potassium phthalimide, DMF, 2h; (iv) N₂H₄·H₂O, NaOH, reflux, 6 h; (v) absolute alcohol, 80 °C, 1.5 h; (vi) glyphosate, deionized water, RT, 30 min.

Herbicidal activities

In this work, target compounds 6a–6d were evaluated for herbicidal activity against Lolium multiflorum Lam. and Brassica campestris. Several commercial herbicides, including PA, 2,4-D, MCPA and glyphosate, were chosen as positive controls. The inhibition rates of those tested compounds were summarized in Tables 1–4, and the toxicity regression equations and IC₅₀ values were shown in Tables 5 and 6. All the synthesized compounds 6a–6d exhibited higher or comparable herbicidal activity than their corresponding DMA or IPA salts. The ammonium salt formed by glyphosate and IPA was named GLYP-IPAM salt. Moreover, those tested compounds possessed remarkable inhibition rates for the root growth but low inhibition rates for the shoot growth of Lolium multiflorum Lam. and Brassica campestris. It can also be observed that the inhibition rates of compounds 5 and 6a–6d were increased with the increase of applications in the range of 0.002 to 1.25 mmol L⁻¹ under certain circumstances. Furthermore, it was found that compounds 6b and 6c containing the Cl atom displayed the strongest herbicidal activity against the root and shoot growth of Lolium multiflorum Lam. and Brassica campestris within the tested concentration range. Impressively, compound 6b had the most potent inhibition efficacy for Brassica campestris root, which maintained an excellent level of effectiveness at a concentration as low as 0.002 mmol L⁻¹, and displayed 100% inhibition rate to the root growth of Brassica campestris at the dosage of 0.020 mmol L⁻¹. When treated at a higher concentration (0.078 mmol L⁻¹), compound 6c also presented a 100% inhibition rate to the root growth of Brassica campestris.

Table 1 Inhibition rates of compounds on the root growth of Lolium multiflorum Lam

Compd	Concentrations (mmol L^−1)
	1.25^a	0.625^a	0.313^a	0.156^a	0.078^a	0.039^a	0.020^a	0.010^a	0.005^a	0.002^a
a The concentration of different longifolene-derived compounds and four ammonium salts solutions (mmol L^−1).b The inhibition rate (%).c The data at this concentration were not determined.
5	100^b	100	100	94.1	70.4	44.0	31.8	23.3	12.9	8.24
6a	100	100	100	95.1	93.4	36.2	9.1	^c	^c	^c
6b	100	100	100	100	100	100	90.0	81.1	39.5	12.7
6c	100	100	100	100	100	100	96.2	89.1	70.1	54.4
6d	100	100	100	98.0	94.7	64.6	11.3	−17.7	−27.9	−24.3
DMA salt of PA	^c	46.7	16.2	13.8	4.1	1.8	0.1	−1.1	−5.3	−5.1
DMA salt of 2,4-D	^c	100	100	98.2	97.2	83.9	70.5	36.1	6.0	2.1
DMA salt of MCPA	^c	99.9	99.8	99.3	97.6	98.4	96.0	90.8	77.5	87.8
GLYP-IPAM salt	^c	88.2	87.9	84.5	72.4	60.9	51.9	50.7	43.9	27.3

---

Table 2 Inhibition rates of compounds on the shoot growth of Lolium multiflorum Lam

Compd	Concentrations (mmol L^−1)
	1.25^a	0.625^a	0.313^a	0.156^a	0.078^a	0.039^a	0.020^a	0.010^a	0.005^a	0.002^a
a The concentration of different longifolene-derived compounds and four ammonium salts solutions (mmol L^−1).b The inhibition rate (%).c The data at this concentration were not determined.
5	100^b	89.8	78.8	61.7	47.7	37.4	28.0	25.6	10.6	1.87
6a	100	100	98.2	59.6	45.0	21.9	0	^c	^c	^c
6b	100	100	100	79.7	64.3	58.1	44.2	33.2	−0.66	−15.8
6c	100	100	100	100	78.6	62.5	53.3	46.3	29.8	24.9
6d	100	100	100	52.8	31.5	30.9	−12.2	−8.68	−12.7	−16.9
DMA salt of PA	^c	16.6	12.4	10.0	1.0	1.0	−4.3	−9.3	−13.1	−10.5
DMA salt of 2,4-D	^c	95.9	91.5	70.0	68.7	53.8	49.4	35.7	13.6	10.8
DMA salt of MCPA	^c	63.1	63.0	61.5	57.9	57.3	56.6	54.1	48.7	35.8
GLYP-IPAM salt	^c	58.0	55.2	42.7	42.3	35.1	26.0	22.0	18.6	14.0

---

Table 3 Inhibition rates of compounds on the root growth of Brassica campestris

Compd	Concentrations (mmol L^−1)
	1.25^a	0.625^a	0.313^a	0.156^a	0.078^a	0.039^a	0.020^a	0.010^a	0.005^a	0.002^a
a The concentration of different longifolene-derived compounds and four ammonium salts solutions (mmol L^−1).b The inhibition rate (%).c The data at this concentration were not determined.
5	100^b	100	100	94.3	60.1	40.4	31.4	22.9	13.6	0.16
6a	100	100	100	100	79.1	31.8	−26.3	^c	^c	^c
6b	100	100	100	100	100	100	100	99.0	97.6	96.3
6c	100	100	100	100	100	98.6	96.0	93.4	90.0	88.0
6d	100	100	100	96.3	90.4	61.6	36.1	11.7	−1.17	0.74
DMA salt of PA	^c	89.0	82.2	67.6	41.7	40.3	27.1	18.5	4.5	−5.5
DMA salt of 2,4-D	^c	92.8	90	88.1	86.3	85.9	83.8	82.8	82.1	81.8
DMA salt of MCPA	^c	100	100	100	100	100	100	100	100	100
GLYP-IPAM salt	^c	79.5	71.2	64.5	63.6	48.2	42.5	24.1	15.7	6.63

---

Table 4 Inhibition rates of compounds on the shoot growth of Brassica campestris

Compd	Concentrations (mmol L^−1)
	1.25^a	0.625^a	0.313^a	0.156^a	0.078^a	0.039^a	0.020^a	0.010^a	0.005^a	0.002^a
a The concentration of different longifolene-derived compounds and four ammonium salts solutions (mmol L^−1).b The inhibition rate (%).c The data at this concentration were not determined.
5	100^b	87.7	76.7	57.5	40.8	35.0	28.2	21.4	19.4	5.44
6a	100	96.5	71.7	54.6	31.2	10.4	9.1	^c	^c	^c
6b	100	100	100	100	96.1	95.1	92.1	91.2	90.0	88.0
6c	100	100	100	100	98.4	97.2	94.3	91.1	87.0	82.6
6d	100	100	100	63.1	49.1	−1.22	−7.30	−2.74	−0.84	−3.59
DMA salt of PA	^c	63.9	50.2	32.9	12.3	12.9	7.4	5.1	−4.5	3.0
DMA salt of 2,4-D	^c	86.8	84.3	82.2	80.1	78.5	76.8	76.3	76.1	71.6
DMA salt of MCPA	^c	93.8	92	87.5	86.6	84.8	82.0	81.1	79.0	77.1
GLYP-IPAM salt	^c	57.9	35.7	32.2	23.3	23.0	22.5	15.6	12.4	8.2

---

Table 5 Toxicity regression equations and IC₅₀ of compounds against Lolium multiflorum Lam.

Compd	Root		Shoot
	Toxicity regression equation	IC50 (mmol L^−1)	Toxicity regression equation	IC50 (mmol L^−1)
5	Y = 2.572 + 1.675x	0.0291	Y = 1.553 + 1.274x	0.0604
	R^2 = 0.929		R^2 = 0.962
6a	Y = 5.291 + 3.894x	0.0438	Y = 3.130 + 3.204x	0.105
	R^2 = 0.919		R^2 = 0.887
6b	Y = 5.084 + 2.281x	0.0059	Y = 2.313 + 1.467x	0.0265
	R^2 = 0.975		R^2 = 0.984
6c	Y = 5.267 + 2.005x	0.0024	Y = 2.446 + 1.288x	0.0126
	R^2 = 0.991		R^2 = 0.977
6d	Y = 6.104 + 4.173x	0.034	Y = 2.534 + 2.458x	0.093
	R^2 = 0.947		R^2 = 0.772
DMA salt of PA	Y = 0.152 + 1.704x	0.814	Y = −0.641 + 2.389x	3.157
	R^2 = 0.96		R^2 = 0.929
DMA salt of 2,4-D	Y = 4.617 + 2.508x	0.014	Y = 1.73 + 1.113x	0.969
	R^2 = 0.976		R^2 = 0.028
DMA salt of MCPA	Y = 3.928 + 1.336x	0.001	Y = 0.654 + 0.452x	0.036
	R^2 = 0.908		R^2 = 0.662
GLYP-IPAM salt	Y = 1.455 + 0.755x	0.012	Y = 0.388 + 0.516x	0.177
	R^2 = 0.966		R^2 = 0.962

---

Table 6 Toxicity regression equations and IC₅₀ of compounds against Brassica campestris

Compd	Root		Shoot
	Toxicity regression equation	IC50 (mmol L^−1)	Toxicity regression equation	IC50 (mmol L^−1)
5	Y = 2.621 + 1.790x	0.0340	Y = 1.391 + 1.156x	0.0630
	R^2 = 0.886		R^2 = 0.951
6a	Y = 6.040 + 4.702x	0.0520	Y = 1.903 + 2.185x	0.1350
	R^2 = 0.982		R^2 = 0.941
6b	Y = 5.552 + 1.486x	0.0002	Y = 2.847 + 0.711x	0.0001
	R^2 = 0.966		R^2 = 0.898
6c	Y = 3.541 + 0.953x	0.0002	Y = 0.3210 + 0.896x	0.0003
	R^2 = 0.947		R^2 = 0.993
6d	Y = 4.048 + 2.605x	0.028	Y = 3.291 + 3.168x	0.092
	R^2 = 0.991		R^2 = 1.000
DMA salt of PA	Y = 1.457 + 1.258x	0.069	Y = 0.458 + 1.085x	0.379
	R^2 = 0.972		R^2 = 0.917
DMA salt of 2,4-D	Y = 0.916 + 0.169x	0.000034	Y = 0.532 + 0.197x	0.002
	R^2 = 0.745		R^2 = 0.685
DMA salt of MCPA	Y = 6.355 + 1.509x	0.000061	Y = 1.139 + 0.099x	0.000147
	R^2 = 0.873		R^2 = 0.923
GLYP-IPAM salt	Y = 1.226 + 0.96x	0.053	Y = 0.107 + 0.589x	0.657
	R^2 = 0.967		R^2 = 0.941

---

For Lolium multiflorum Lam., the herbicidal activity of compounds 6b and 6c were higher than that of compounds 5, 6a and 6d. Their inhibition rates for root growth were exceeded 90% at the concentration of 0.02 mmol L⁻¹ and reached 100% at 0.039 mmol L⁻¹. The inhibition rate of compound 6c on shoot growth were 100% when treated at 0.156 mmol L⁻¹. It is worth noting that almost all the synthesized compounds completely inhibited Lolium multiflorum Lam. root and shoot growth at 0.313 mmol L⁻¹ (Fig. S17 and S19). On the other hand, the order of the herbicidal activity against the root growth of Brassica campestris was 6b > 6c > 6a > 6d > 5, while the order of the herbicidal activity against the shoot growth was 6c ≥ 6b > 6d > 6a > 5. Compound 6c exhibited a slightly elevated control over Brassica campestris shoot growth in comparison to compound 6b when applied at the concentration ranging from 0.02 to 0.078 mmol L⁻¹. However, the inhibitory effects of compounds 6b and 6c against the shoot growth of Brassica campestris were the same with 100% inhibition at the concentration of 0.156 mmol L⁻¹ (Fig. S18 and S20). Although compound 6c demonstrated excellent inhibition rates for root growth, its older DMA salt formulation exhibited superior herbicidal performance, which showed complete inhibition even at 0.002 mmol L⁻¹. The inhibition rate of compound 6a for Brassica campestris root growth was higher than that for shoot growth, at the concentration of 0.156 mmol L⁻¹, the inhibition rates against shoot growth were 54.6%, but against root growth reached 100%. Additionally, compound 6d showed 100% of control efficacy both on Brassica campestris root and shoot growth at the dosage of 0.313 mmol L⁻¹ (Fig. S21–S22).

As shown in Table 5, the IC₅₀ values of compounds 6a–6d against Lolium multiflorum Lam. root and shoot growth were 0.0024–0.0438 and 0.0126–0.1050 mmol L⁻¹, respectively. Among them, compounds 6b and 6c possessed much lower IC₅₀ values than that of compound 5 (IC₅₀ values of root and shoot growth were 0.0291 and 0.0604 mmol L⁻¹, respectively). It seemed that the IC₅₀ values of compounds 6a–6c against Lolium multiflorum Lam. root and shoot growth were lower than that of their corresponding DMA salts. The IC₅₀ value of compound 6d against Lolium multiflorum Lam. root growth was lower than that of GLYP-IPAM salt, but higher than GLYP-IPAM salt against Lolium multiflorum Lam. shoot growth. According to Fig. 1, the herbicidal activity of compound 6a against root growth of Lolium multiflorum Lam. was more than 18 times higher than that of DMA salt of PA, and the herbicidal activity of compound 6b was 137.3% higher than that of DMA salt of 2,4-D against Lolium multiflorum Lam. root growth. Similarly, compounds 6a and 6b showed 2906.7% and 3556.6% higher herbicidal activity against Lolium multiflorum Lam. shoot growth than their corresponding DMA salts, respectively. Moreover, compounds 6c and 6d exhibited 185.7% and 90.3% higher herbicidal activity against Lolium multiflorum Lam. shoot growth than DMA salt of MCPA and GLYP-IPAM salt, respectively.

Fig. 1 Herbicidal effects of compounds against the root growth (A) and shoot growth (B) of Lolium multiflorum Lam. compared to that of their corresponding DMA/IPA salts.

From Table 6, the IC₅₀ values of compounds 6a–6d against Brassica campestris root and shoot growth were 0.0002–0.052 mmol L⁻¹ and 0.0001–0.135 mmol L⁻¹, respectively. In particular, compounds 6b and 6c with the lowest IC₅₀ values (around 0.0002 mmol L⁻¹) were more favourable to herbicidal activity than compounds 5, 5a and 5d. It is important to note that compound 6b displayed 200 times higher herbicidal activity against Brassica campestris shoot growth than DMA salt of 2,4-D. Moreover, compounds 6a and 6d both had lower IC₅₀ values than their DMA/IPA salt. It can be seen from Fig. 2 that compounds 6a and 6d presented 180.7% and 614.1% higher herbicidal activity against Brassica campestris shoot growth than their corresponding salts, and their herbicidal activities against Brassica campestris root growth were 32.7% and 89.3% higher than that of DMA salt of PA and GLYP-IPAM salt, respectively. For compounds 6b and 6c containing one or two chlorine atoms, despite they displayed higher inhibition rates and lower IC₅₀ values than their corresponding DMA salts for the root and shoot growth of Brassica campestris in most cases, only compound 6b had a relatively lower IC₅₀ value against Brassica campestris shoot growth than DMA salt of 2,4-D.

Fig. 2 Herbicidal effects of compounds against the root growth (A) and shoot growth (B) of Brassica campestris compared to that of their corresponding DMA/IPA salts.

Herbicidal spectrum of compound 6c

Based on the results, the IC₅₀ values of compound 6c against Lolium multiflorum Lam. and Brassica campestris were almost the lowest among the prepared compounds, compound 6c was further examined for its inhibitory effects on six different types of weeds (Table 7). It is shown that compound 6c had different inhibitory activities in the concentration range of 0.0001–0.039 mmol L⁻¹. Compound 6c was more sensitive on the root growth of those weeds than that on the shoot growth, and its inhibition rates were also related to the applied concentration. At the concentration of 0.010 mmol L⁻¹, compound 6c displayed over 80% inhibition against the root growth of Setaria viridis, Eleusine indica, rice, Portulaca oleracea and clover, whereas, compound 6c only exhibited less than 30% inhibition against the shoot growth of Eleusine indica, rice and Portulaca oleracea at the same dosage. When the concentration was 0.039 mmol L⁻¹, the inhibition rates of compound 6c on the root growth of two monocotyledonous weeds (Setaria viridis and rice) were 100%, while the inhibition rate on the root growth of Eleusine indica was 91.4%. In terms of dicotyledonous weeds, compound 6c displayed > 95% inhibition against Portulaca oleracea root growth and clover root and shoot growth when treated at 0.039 mmol L⁻¹, whereas compound 6c only displayed < 85% inhibition against Medicago sativa L. root and shoot growth and Portulaca oleracea shoot growth at the same concentration. Additionally, compound 6c had much lower IC₅₀ values against the root growth of the tested weeds than that against the shoot growth, as the IC₅₀ values of root and shoot growth were 0.000085–0.000612 and 0.00188–1.31 mmol L⁻¹, respectively (Table 8). In general, compound 6c showed good broad-spectrum weed control against the weeds tested, and the control effect on the root growth was better than that on the shoot growth, especially for monocotyledonous weeds.

Table 7 Inhibition rates of compound 6c against different plants

Concentrations (mmol L^−1)	Setaria viridis		Eleusine indica		Rice		Portulaca oleracea		Medicago sativa L.		Clover
	Root	Shoot	Root	Shoot	Root	Shoot	Root	Shoot	Root	Shoot	Root	Shoot
0.0001	45.2	−9.07	39.6	3.98	72.4	11.8	38.4	−7.35	19.8	18.5	47.1	7.53
0.0006	58.1	8.27	43.7	21.1	84.8	12.5	41.6	−5.39	43.8	24.1	67.3	10.3
0.001	74.2	9.33	63.5	21.9	94.1	15.1	67.2	7.35	58.3	37.7	76.9	33.6
0.002	77.4	24.0	62.9	20.3	95.9	13.2	84.8	27.5	66.7	63.5	79.8	57.5
0.010	83.9	50.4	85.3	25.9	100	−1.32	95.6	26.9	63.5	81.5	85.7	81.5
0.039	100	71.7	91.4	43.4	100	32.2	96.8	77.0	76.0	84.8	95.2	95.9

---

Table 8 Toxicity regression equations and IC₅₀ of compound 6c against different plants

Plants	Root		Shoot
	Toxicity regression equation	IC50 (mmol L^−1)	Toxicity regression equation	IC50 (mmol L^−1)
Setaria viridis	Y = 2.498 + 0.649x	0.000142	Y = 2.304 + 1.179x	0.0111
	R^2 = 0.947		R^2 = 0.979
Eleusine indica	Y = 2.486 + 0.774x	0.000612	Y = 0.510 + 0.493x	0.0922
	R^2 = 0.933		R^2 = 0.897
Rice	Y = 5.448 + 1.338x	0.000085	Y = −0.40 + 0.343x	1.31
	R^2 = 0.882		R^2 = 0.846
Portulaca oleracea	Y = 2.935 + 0.817x	0.000257	Y = 2.351 + 1.146x	0.00887
	R^2 = 0.894		R^2 = 0.910
Medicago sativa L.	Y = 1.253 + 0.389x	0.000602	Y = 2.531 + 0.929x	0.00188
	R^2 = 0.784		R^2 = 0.931
Clover	Y = 2.912 + 0.805x	0.000241	Y = 3.633 + 1.376x	0.00229
	R^2 = 0.921		R^2 = 0.944

---

Conclusions

In summary, a library of novel ammonium phenoxyacetates and glyphosate based on longifolene-derived primary amine were synthesized and characterized. Herbicidal activity evaluation showed that the majority of the synthesized compounds performed better than their corresponding herbicide-amine salts. Compounds 6b and 6c with the Cl atom displayed the most potent herbicidal efficacy for Lolium multiflorum Lam. and Brassica campestris, Lolium multiflorum Lam. root growth and Brassica campestris shoot growth were both completely inhibited when their treated concentrations were 0.039 and 0.156 mmol L⁻¹. Moreover, the root growth of Brassica campestris was completely inhibited by compounds 6b and 6c at the concentrations of 0.02 and 0.078 mmol L⁻¹, respectively. Furthermore, compound 6c was found to have a broad spectrum of weed control and displayed a good herbicidal effect on the root growth of the experimental weeds at 0.039 mmol L⁻¹. Those findings indicated that compounds 6b and 6c would have great potential to serve as high-performance botanical herbicide candidates used at low doses.

Experimental section

Material

Heavy turpentine was obtained commercially from Guangxi Wusong Pine Chemicals Group Co., Ltd (Guangxi, China). Longifolene (compound 1) with a boiling point of 252–254 °C was synthesized from heavy turpentine based on a previously published method.²⁷ Compounds 2 and 3 were prepared following the procedure reported in our previous work.^24,25 Compound 3 was obtained by the Prins and halogenation reaction from compound 1,²⁴ and compound 4 was prepared by using potassium phthalimide.^22,23 PA, 2,4-D, MCPA, glyphosate, acetonitrile and DMF were purchased from Shanghai Aladdin Biochemical Technology Co., Ltd (Shanghai, China). Absolute alcohol, sodium hydroxide (NaOH), Tween 80, deuterated trichloromethane (CDCl₃), deuterated methanol (CD₃OD) and ethyl acetate were purchased from Tianjin Damao Chemical Reagent Factory (Tianjin, China). Lolium multiflorum Lam. and Brassica campestris were purchased from Barenbrug International Co., Ltd (Tianjin, China), and Shenzhen Aoxin Libao Industrial Co., Ltd (Shenzhen, China), respectively. Different formulations of DMA salts and GLYP-IPAM salt were prepared through the neutralization reaction in our labs according to the procedure described in patents.^28,29

Synthesis of compound 5

Compound 5 was synthesized in accordance with the procedure in our previous study.^22,23 To a mixture of compound 4 (2.181 g, 6 mmol) and 80% N₂H₄·H₂O (8.635 g, 138 mmol), 5 g of 10% aqueous NaOH solution was added. Then, the reaction mixture was refluxed for 6 h at 120 °C. After cooling, the mixture was extracted with petroleum ether (3 × 10 mL). The combined organic phase was washed with distilled water (3 × 10 mL) and dried over anhydrous Na₂SO₄ and then evaporated the solvent in a vacuum to gain compound 5 as a pale-yellow oily liquid with yield of 94.1%. The boiling point of compound 5 was 288–290 °C.

Synthesis of compounds 6a–6c

The preparation of compounds 6b and 6c followed the typical procedure described below for the preparation of compound 6a. PA (0.76 g, 5 mmol) was added to a three-necked flask with absolute alcohol (15 mL), respectively. The mixture was stirred for 15 min at room temperature, and a mixture of compound 5 (1.17 g, 5 mmol) in absolute alcohol (8 mL) was added dropwise. Then, the reaction mixture was heated to 80 °C and stirred for 1.5 h. After cooling to room temperature, the mixture was evaporated to remove the volatile compounds. The crude produce was washed with absolute alcohol and ethyl acetate (5 : 1, v/v) to obtain compound 6a.

(E)-2-(4,8,8-trimethyldecahydro-1,4-methanoazulen-9-ylidene)ethan-1-aminium 2-phenoxyacetate (6a). Yield: 81.7%; white powder; melting point (mp), 133.1–134.1 °C. FT-IR (KBr) v (cm⁻¹): 3115.16, 2958.80, 2929.87, 1685.79, 1639.49, 1597.06, 1571.99, 1544.98, 1496.76, 1473.62, 1456.26, 1415.75, 1375.25, 1338.60, 1288.45, 1236.37, 1174.65, 1084.00, 1055.06, 937.40, 837.11, 748.38, 707.88, 688.59. ¹H NMR (400 MHz, CD₃OD) δ (ppm), 7.35–7.16 (m, 2H), 7.02–6.82 (m, 3H), 5.04 (t, ¹H, J = 7.1 Hz), 4.39 (s, 2H), 3.67–3.49 (m, 2H), 2.97 (d, ¹H, J = 4.9 Hz), 2.17 (d, ¹H, J = 3.5 Hz), 1.81 (tdd, ¹H, J = 12.0, 5.0, 3.3 Hz), 1.75–1.59 (m, 3H), 1.58–1.41 (m, 4H), 1.22–1.07 (m, 2H), 1.04 (s, 3H), 1.00 (s, 3H), 0.93 (s, 3H). ¹³C NMR (100 MHz, CDCl₃) δ (ppm), 175.01, 164.52, 158.55, 129.51, 120.89, 114.69, 107.49, 67.64, 62.35, 44.77, 44.43, 43.23, 42.32, 38.33, 36.49, 33.48, 30.94, 30.19, 30.04, 28.66, 25.45, 20.86. HRMS (ESI) m/z [M–C₈H₇O₃–NH₃]⁺ calculated for C₁₆H₂₅ 217.1956, found 217.1960; [M-C₁₆H₂₈N]⁻ calculated for C₈H₇O₃ 151.0395, found 151.0397.

(E)-2-(4,8,8-trimethyldecahydro-1,4-methanoazulen-9-ylidene)ethan-1-aminium 2-(2,4-dichlorophenoxy)acetate (6b). Yield: 90.6%; white powder; mp 145.9–146.7 °C. FT-IR (KBr) v (cm⁻¹): 3113.11, 3051.39, 2947.23, 2866.22, 1635.64, 1593.20, 1529.55, 1477.47, 1402.25, 1284.59, 1263.38, 1232.51, 1105.21, 1068.56, 1039.63, 908.47, 869.90, 835.18, 804.32, 719.45, 646.15. ¹H NMR (400 MHz, CDCl₃) δ (ppm), 7.33 (d, ¹H, J = 2.5 Hz), 7.13 (dd, ¹H, J = 8.8, 2.5 Hz), 6.81 (d, ¹H, J = 8.9 Hz), 4.89 (t, ¹H, J = 7.0 Hz), 4.42 (s, 2H), 3.41 (dd, 2H, J = 6.9, 3.1 Hz), 2.73 (d, ¹H, J = 4.6 Hz), 2.05 (d, ¹H, J = 3.1 Hz), 1.66 (s, ¹H), 1.59–1.47 (m, 3H), 1.36 (ddd, 5H, J = 24.5, 17.5, 10.9 Hz), 1.00 (ddd, 2H, J = 17.2, 9.8, 5.1 Hz), 0.92 (s, 3H), 0.88 (s, 3H), 0.81 (s, 3H). ¹³C NMR (100 MHz, CDCl₃) δ (ppm), 173.94, 165.29, 153.07, 129.97, 127.72, 125.97, 123.17, 114.63, 107.01, 68.66, 62.34, 44.70, 44.46, 43.25, 42.35, 38.48, 36.50, 33.46, 30.89, 30.15, 30.00, 28.72, 25.41, 20.85. HRMS (ESI) m/z [M–C₈H₅O₃Cl₂–NH₃]⁺ calculated for C₁₆H₂₅ 217.1956, found 217.1956; [M–C₁₆H₂₈N]⁻ calculated for C₈H₅O₃Cl₂ 218.9616, found 218.9608.

(E)-2-(4,8,8-trimethyldecahydro-1,4-methanoazulen-9-ylidene)ethan-1-aminium 2-(4-chloro-2-methylphenoxy)acetate (6c). Yield: 73.4%; white powder; mp 155.8–156.5 °C. FT-IR (KBr) v (cm⁻¹): 3057.17, 2956.88, 2924.09, 2912.51, 2879.72, 2866.22, 1635.64, 1593.20, 1560.41, 1490.98, 1456.26, 1404.18, 1371.39, 1298.09, 1257.59, 1228.66, 1190.08, 1134.14, 1060.85, 877.61, 804.32, 705.95, 646.15. ¹H NMR (400 MHz, CDCl₃) δ (ppm), 7.08 (d, ¹H, J = 2.0 Hz), 7.04 (dd, 1H, J = 8.6, 2.4 Hz), 6.66 (d, ¹H, J = 8.6 Hz), 4.86 (s, ¹H), 4.37 (s, 2H), 3.26 (t, 2H, J = 7.4 Hz), 2.70 (d, ¹H, J = 4.5 Hz), 2.22 (s, 3H), 2.06 (d, ¹H, J = 2.7 Hz), 1.76–1.62 (m, ¹H), 1.59–1.48 (m, 3H), 1.35 (ddd, 5H, J = 30.1, 22.2, 8.8 Hz), 1.09–0.98 (m, 2H), 0.92 (s, 3H), 0.89 (s, 3H), 0.83 (s, 3H). ¹³C NMR (100 MHz, CDCl₃) δ (ppm), 175.14, 165.32, 155.62, 130.56, 128.66, 126.41, 125.41, 112.91, 106.97, 68.34, 62.34, 44.69, 44.46, 43.25, 42.33, 38.32, 36.53, 33.47, 30.93, 30.16, 30.01, 28.71, 25.41, 20.85, 16.34. HRMS (ESI) m/z [M–C₉H₈O₃Cl]⁺ calculated for C₁₆H₂₈N 234.2222, found 234.2222; [M–C₁₆H₂₈N]⁻ calculated for C₉H₈O₃Cl 199.0162, found 199.0153.

Synthesis of compounds 6d

Compound 5 (0.234 g, 1 mmol) and glyphosate (0.169 g, 1 mmol) were added to 10 mL of deionized water at room temperature for 30 min to obtain a homogenous solution. After that, acetonitrile was added dropwise into the solution. When milky turbid appeared in the solution system, the precipitate was collected by centrifugation. The collected solid was then processed by vacuum drying to obtain compound 6d.

(E)-2-(4,8,8-trimethyldecahydro-1,4-methanoazulen-9-ylidene)ethan-1-aminium (phosphonomethyl)glycinate (6d). Yield: 82.2%; white powder; mp 168.2–182.3 °C. FT-IR (KBr) v (cm⁻¹): 3397, 3331, 2955, 2851, 1738, 1640, 1157, 1025, 907, 748. ¹H NMR (400 MHz, D₂O) δ (ppm), 4.98–5.02 (t, ¹H, H-2′, J = 8.0 Hz), 3.69 (s, 2H, H-2′′), 3.57–3.65 (m, 2H, H-1′), 3.15–3.18 (d, J = 12.0 Hz, 2H, H-4′′), 2.92–2.93 (d, J = 4.0 Hz, ¹H, NH), 2.12–2.13 (d, ¹H, H-1), 1.69–1.73 (m, ¹H, H-3eq), 1.57–1.63 (m, 3H, H-3ax, H-6eq, H-2eq), 1.49–1.61 (m, 2H, H-6ax, H-2ax), 1.31–1.44 (m, 3H, H-3a, H-5eq, H-7eq), 1.09–1.14 (m, ¹H, H-7ax), 0.98–1.04 (m, 2H, H-5ax, H-8a), 0.93 (s, 3H, 8-CH₃), 0.90 (s, 3H, 8-CH₃), 0.82 (s, 3H, 4-CH₃). ¹³C NMR (100 MHz, D₂O) δ (ppm), 170.81, 167.89, 105.24, 61.93, 50.43, 44.59, 44.34, 44.18, 43.22, 42.92, 42.17, 38.46, 36.19, 32.81, 30.14, 29.37, 28.41, 24.89, 23.11. HRMS (SI) m/z [M–NH₃]⁺ calculated for C₁₆H₂₅ 217.1956, found 217.1958; [M–C₁₆H₂₈N]⁻ calculated for C₃H₇NO₅P⁻ 168.0067, found 165.0059.

Characterization

Structural elucidation of all the synthesized compounds were achieved by proton and carbon NMR spectroscopy. Proton (¹H) and carbon (¹³C) NMR spectrum of the products were recorded on a Bruker Ascend™ 400 MHz spectrometer (Bruker Co., Ltd, Fällanden, Switzerland) with CDCl₃ or CD₃OD as the solvent and tetramethylsilane (TMS) as the internal standard. The FT-IR spectra were recorded using a Magna-IR 550 (II) Fourier transform infrared spectrometer (Nicolet Co., Ltd, USA) with the use of potassium bromide pellets. All the spectra were captured in the 400–4000 cm⁻¹ region, at a resolution of 4 and 64 scans per sample. Before analyzing each sample, the air was used as a reference. The HRMS spectral analysis was performed on an Agilent mass spectrometer (Agilent Technologies, Palo Alto, USA) under electron spray ionization. The melting point was measured by Beijing Taike point apparatus (X-4) and was uncorrected, and the boiling point was determined through a simple distillation method.

Herbicidal activity evaluation

According to the method reported in the literature, herbicidal activity was evaluated with three replicates per treatment.^30,31 Lolium multiflorum Lam. and Brassica campestris were selected to assess the preliminary herbicidal activity of compounds 6a–6d. The target compounds (2 mmol) were dissolved respectively in 0.5 mL DMF in a 100 mL volumetric flask to create a mother solution of 20 mmol L⁻¹. These solutions were further diluted using 1% Tween-80 and distilled water to achieve the desired concentration. The same amount of distilled water including DMF and Tween-80 was served as the blank control. The seeds were steeped in 2% sodium hypochlorite solution for 5 min, washed with distilled water three times, and soaked in distilled water for 15 h. Afterwards, the treated seeds were placed in Petri dishes (9 cm in diameter) and incubated at 25 °C for 24 h. Ten seeds were chosen to each Petri dishes (9 cm in diameter) containing two pieces of filter paper and 10 mL test solution, incubated in dark for 3 days at 25 °C. The root or shoot length of each seed was measured. The results were expressed as the average of three independent experiments. The inhibition rate of the root or shoot growth of tested seeds was calculated by equation below, the root or shoot growth of tested seeds less than 1 mm would not take into consideration.
where P is the inhibition rate of the root or shoot growth, L₀ is the average root or shoot length of the blank control group in each test, and L is the average root or shoot length of the group treated with the test compounds.

Weed spectrum testing

Further the herbicidal activity study of the compound with the most significant herbicidal activity against Lolium multiflorum Lam. and Brassica campestris was also performed for several representative monocotyledonous and dicotyledonous weeds (Setaria viridis, Eleusine indica, rice, Portulaca oleracea, Medicago sativa L., and clover). The test was studied using the same method used for the evaluation of herbicidal activity.

Author contributions

Synthesis and characterization, Y. Q. H., P. P. L.; herbicidal activity evaluation, F. G. Y., Z. Q. G., J. X. W., Z. Q. Z.; writing, review and editing, Y. Q. H, H. Y. L., D. Z. H., Y. F. All authors have read and agree to the published version of the manuscript.

Conflicts of interest

There are no conflicts to declare.

Data availability

The data supporting this article have been included as part of the supplementary information (SI). Supplementary information: IR, ¹H NMR, ¹³C NMR, and HRMS spectra for all compounds, and herbicidal effects of compounds against tested weeds. See DOI: https://doi.org/10.1039/d5ra05630f.

Acknowledgements

This work was supported by Guangxi Science and Technology Major Program (no. AA24206021), the Natural Science Foundation of Guangxi Zhuang Autonomous Region (no. 2017GXNSFAA198027), and Guangxi College Student Innovation and Entrepreneurship Training Program (no. S202510608224).

Notes and references

1. L. P. Gianessi, Pest Manage. Sci., 2013, 69, 1099–1105.
2. S. M. Sharkey, A. M. Hartig, A. J. Dang, A. Chatterjee, B. J. Williams and K. M. Parker, Environ. Sci. Technol., 2022, 56, 13644–13653.
3. A. Syguda, M. Wojcieszak, K. Materna, M. Wozniak-Karczewska, A. Parus, Ł. Ł awnicazk and Ł. Chrzanowski, ACS Sustainable Chem. Eng., 2020, 8, 13584–14594.
4. Y. Hamisu, J. Auta and B. S. Bawa, FUDMA J. Sci., 2024, 8, 177–185.
5. T. C. Mueller, L. E. Steckel and A. Shekoofa, Weed Technol., 2022, 36, 462–467.
6. A. Qurratu and A. Reehan, Int. J. Appl. Eng. Res., 2016, 11, 9946–9955.
7. U. Umiyati, D. Kurniadie, I, R. D. Anjarsari and M. M. Triazahra, Res. Crop., 2025, 26, 360–367.
8. T. P. Boyle, Environ. Pollut., Ser. A, 1980, 21, 35–49.
9. J. N. Smith, K. C. Barsanti, H. R. Friedli, M. Ehn, M. Kulmala, D. R. Collins, J. H. Scheckman, B. J. Williams and P. H. McMurry, Proc. Natl. Acad. Sci. U. S. A., 2010, 107, 6634–6639.
10. U. Lohmann and J. Feichter, Atmos. Chem. Phys., 2005, 5, 715–737.
11. T. M. Pan, Y. X. Geng, J. Hao, X. H. He, J. Li, Y. Q. Gao, S. B. Shang and Z. Q. Song, J. Agric. Food Chem., 2021, 69, 12985–12993.
12. T. C. Sparks, D. R. Hahn and N. V. Garizi, Pest Manage. Sci., 2017, 73, 700–715.
13. Y. Cao, R. Zhang, W. Liu, G. Zhao, W. Niu, J. Guo, M. Xian and H. Liu, Sci. Rep., 2019, 9, 95.
14. A. Mukai, K. Takahashi and T. Ashitani, J. Wood Sci., 2017, 63, 360–368.
15. S. Y. Zhao, G. S. Lin, W. G. Duan, Q. N. Zhang, Y. L. Huang and F. H. Lei, ACS Omega, 2021, 6, 9104–9111.
16. X. P. Zhu, G. S. Lin, W. G. Duan, Q. M. Li, F. Y. Li and S. Z. Lu, Molecules, 2020, 25, 986–998.
17. A. Mukai, K. Takahashi and T. Ashitani, Eur. J. Wood Wood Prod., 2017, 76, 1079–1082.
18. Q. N. Zhang, G. S. Lin, W. G. Duan, S. Y. Zhao, J. M. He and F. H. Lei, ChemistrySelect, 2021, 6, 4515–4520.
19. H. L. Lan, X. P. Zhu, G. S. Lin, W. G. Duan, Y. C. Cui, F. Y. Li and D. P. Li, Chem. Biodiversity, 2023, 20, e202201163.
20. X. J. Wang, Z. Wang, J. Han, S. H. Su, Y. X. Gong, Y. Zhang, N. H. Tan, J. Wang and L. Feng, J. Agric. Food Chem., 2024, 72, 2598–2611.
21. M. Rostaei, S. Fallah, Z. Lorigooini and A. A. Surki, Ind. Crops Prod., 2018, 125, 622–629.
22. P. P. Lu, Y. Q. Huang, Y. H. Gan, L. J. Huang, H. Ge and D. Z. Huang, Chem. Ind. For. Prod., 2023, 43, 8–16.
23. Y. Q. Huang, P. P. Lu, H. Y. Lan, D. Z. Huang, B. Y. Liu, Q. Z. Wei, L. J. Huang and Y. H. Gan, J. Agric. Food Chem., 2024, 72, 17219–17228.
24. H. Y. Lan, Q. Y. Li, D. Z. Huang, L. Lin, Z. Y. Li and Y. Zhou, Fine Chem., 2018, 35, 1255–1260.
25. H. Y. Lan, S. C. Huo, D. Z. Huang and Y. Z. Lu, Chin. J. Struct. Chem., 2020, 39, 546–1552.
26. J. Masuda, S. Kondo, Y. Mastsumoto and M. Yamanaka, ChemistrySelect, 2018, 3, 6112–6115.
27. D. Z. Huang, H. Y. Lan and Z. Y. Zhao, Fine Chem., 2016, 33, 674–679+692.
28. F. D. Jones, 390941A, US Pat., 2, 1945.
29. Y. Y. Chen, CN104031085, Chinese Pat., 2014.
30. H. M. Zhang, Y. X. Chen, S. C. Xu, J. Wang, H. H. Dong, Z. D. Zhao and J. X. Jiang, RSC Adv., 2021, 11, 27207.
31. Y. Q. Huang, M. M. Zhu, H. Y. Lan, D. Z. Huang, M. N. Lan and Y. N. Huang, Chem. Biodiversity, 2024, 21, e202301867.

---